On the basis of the three-dimensional structure of full-length HSA, tHSA contains the primary binding sites for heme and warfa-rin, and the secondary ibuprofen-binding site Fig.. Binding
Trang 1truncated recombinant human serum albumin – an optical and NMR spectroscopic investigation
Gabriella Fanali1, Giorgio Pariani1, Paolo Ascenzi2,3and Mauro Fasano1
1 Dipartimento di Biologia Strutturale e Funzionale, Universita` dell’Insubria, Busto Arsizio, Italy
2 Istituto Nazionale per le Malattie Infettive IRCCS ‘Lazzaro Spallanzani’, Roma, Italy
3 Laboratorio Interdisciplinare di Microscopia Elettronica, Universita` Roma Tre, Roma, Italy
Human serum albumin (HSA), the most prominent
protein in plasma, is best known for its exceptional
ligand-binding capacity, the most strongly bound
com-pounds being hydrophobic organic anions of medium
size, long-chain fatty acids, heme, and bilirubin
More-over, HSA abundance (its concentration being
45 mgÆmL)1in serum of healthy human adults) makes
it an important determinant of the pharmacokinetic
behavior of many drugs HSA also accounts for most
of the antioxidant capacity of human serum
Further-more, HSA participates in heme iron reuptake
follow-ing hemolytic events, acts as an NO depot, and
displays (pseudo)enzymatic properties [1,2]
The amino acid sequence of HSA shows three homologous domains, probably arising from divergent evolution of a degenerated ancestral gene followed by
a fusion event Terminal regions of sequential domains contribute to the formation of interdomain helices linking domain I to domain II, and domain II to domain III, respectively On the other hand, each domain is known to be composed of two separate sub-domains (named A and B), connected by a random coil The multidomain structural organization of HSA provides a variety of ligand-binding sites (Fig 1) [1–5] Among them, two main drug-binding regions have been identified, and named as Sudlow’s sites [6]
Keywords
human serum albumin; ibuprofen; nuclear
magnetic relaxation dispersion; truncated
human serum albumin; warfarin
Correspondence
M Fasano, Dipartimento di Biologia
Strutturale e Funzionale, Universita`
dell’Insubria, Via A da Giussano, 12,
I-21052 Busto Arsizio (VA), Italy
Fax: +39 0331 339459
Tel: +39 0331 339450
E-mail: mauro.fasano@uninsubria.it
(Received 10 October 2008, revised 29
December 2008, accepted 6 February 2009)
doi:10.1111/j.1742-4658.2009.06952.x
Human serum albumin (HSA) is known for its exceptional ligand-binding capacity; indeed, its modular domain organization provides a variety of ligand-binding sites Its flexible modular structure involves more than the immediate vicinity of the binding site(s), affecting the ligand-binding prop-erties of the whole protein Here, biochemical characterization by
1H-NMR relaxometry and optical spectroscopy of a truncated form of HSA (tHSA) encompassing domains I and II (Asp1–Glu382) is reported Removal of the C-terminal domain III results in a number of contacts that involve domain I (containing the heme site) and domain II (containing the warfarin site) being lost; however, the allosteric linkage between heme and warfarin sites is maintained tHSA shows a nuclear magnetic relaxation dis-persion profile similar to that of HSA, and displays increased affinity for ibuprofen, warfarin, and heme, suggesting that the fold is preserved More-over, the allosteric properties that make HSA a peculiar monomeric protein and account for the regulation of ligand-binding modes by heterotropic interactions are maintained after removal of domain III Therefore, tHSA
is a valuable model with which to investigate allosteric properties of HSA, allowing independent analysis of the linkages between different drug-binding sites
Abbreviations
HSA, human serum albumin; NMRD, nuclear magnetic relaxation dispersion; tHSA, truncated human serum albumin; ZFS, zero field splitting.
Trang 2Ibuprofen, a nonsteroidal anti-inflammatory agent,
and warfarin, a coumarinic anticoagulant drug, are
considered to be stereotypical ligands for Sudlow’s
site II and Sudlow’s site I, respectively
Warfarin binds to Sudlow’s site I with
Kd= 3.0· 10)6m, in a pocket formed by the packing
of all six helices of subdomain IIA [3,7–9] The
interac-tion between warfarin and HSA appears to be
domi-nated by hydrophobic contacts, although specific
electrostatic interactions are observed Ibuprofen binds
primarily to Sudlow’s site II, with Kd= 1.8· 10)6m
[3,10,11] Site II is composed of all six helices of
sub-domain IIIA, and it is topologically similar to site I,
with the exception that it may accommodate two fatty
acid anions A secondary ibuprofen site has been
located at the interface between subdomains IIA and
IIB [12] Moreover, multiple recognition sites for drug,
fatty acid and hormone binding to HSA have also
been identified [1,2,8,12,13]
Heme endows HSA with peculiar optical and
mag-netic spectroscopic properties, which can be used to
investigate ligand-dependent and pH-dependent
struc-tural properties [9,14–19] Heme binds to HSA in a
D-shaped cavity limited by Tyr138 and Tyr161, which
provide p–p stacking interactions with the porphyrin;
Tyr161 supplies a donor oxygen to the ferric heme
iron, forming a pentacoordinate high-spin system [20]
Heme propionates point towards the interface between
domains I and III, and are stabilized by salt bridges
with Arg114 and Lys190 residues [21,22] Interestingly,
the heme site of HSA has a low affinity for long-chain
and medium-chain fatty acids, suggesting that its
geometry has evolved to specifically bind to the heme
[23,24]
The conformational adaptability of HSA involves more than the immediate vicinity of the binding site(s), affecting both the structure and the ligand-binding properties of the whole HSA molecule, which displays ligand-dependent allosteric conformational transi-tion(s) [1,2] Heme regulates drug binding to Sudlow’s site I by heterotropic interactions Indeed, the affinity
of Fe(III)heme for HSA decreases by about one order
of magnitude upon drug binding, and accordingly Fe(III)heme binding to HSA decreases drug affinity to the same extent Therefore, drugs that bind to Sud-low’s site I (e.g warfarin) act as allosteric effectors for Fe(III)heme association, and vice versa [9,18,25–29] Also, the heme cleft and the secondary ibuprofen site are allosterically coupled [18,23] Furthermore, drugs allosterically modulate heme–HSA reactivity [20,30] HSA also undergoes pH-induced conformational transitions Between pH 2.7 and pH 4.3, HSA shows a fast-migrating (F) form, characterized by a dramatic increase in viscosity, low solubility, and a significant loss of the a-helical content Between pH 4.3 and
pH 8.0, and in the absence of allosteric effectors, HSA displays the neutral (N) form, which is characterized
by a ‘heart-shaped’ structure At pH values > 8, and
in the absence of ligands, HSA changes conformation
to the basic (B) form, which displays increased affinity for some ligands [1–3,9,14–16,19,31–33]
Few years ago, five recombinant HSA fragments were prepared and characterized, in order to identify the protein region containing the warfarin primary binding site [7,34] Here, we report a thorough bio-chemical characterization, including Fe(III)heme-bind-ing properties, of a truncated form of HSA (tHSA) encompassing residues Asp1–Glu382, which corre-spond to domains I and II On the basis of the three-dimensional structure of full-length HSA, tHSA contains the primary binding sites for heme and warfa-rin, and the secondary ibuprofen-binding site (Fig 1)
Results and Discussion
Dynamics and hydration of tHSA Figure 2 shows the nuclear magnetic relaxation disper-sion (NMRD) profiles of 1.0· 10)3m HSA and tHSA solutions at pH 7.0 and 25 C The data shown here have been analyzed using Eqn (1), and are consistent with a molecular correlation time sc of 20 ± 1 ns for tHSA, which appears reasonable in comparison to
sc= 48 ± 2 ns obtained for full-length HSA under the same experimental conditions (Table 1) Indeed, the molecular correlation time is dependent on the molecular mass of the molecule A systematic analysis
Fig 1 Heme (Protein Data Bank entry: 1O9X [22]), warfarin
(Pro-tein Data Bank entry: 2BXD [12]), and ibuprofen (Pro(Pro-tein Data Bank
entry: 2BXG [12]) modes of binding to HSA Domains I and II are
rendered as blue and orange ribbons, respectively Domain III,
which has been removed in tHSA, is rendered as pale red ribbons.
Heme, warfarin and ibuprofen are rendered in black as ball and
stick.
Trang 3of a number of proteins with different sizes indicates
that such a value could be expected for a 44 kDa
pro-tein [35]; therefore, solution dynamics indicate that
tHSA is not aggregated or misfolded
The analysis of the amplitude of the NMRD profile
[i.e b in Eqn (1)] can provide quantitative
informa-tion on the number of water molecules contributing to
the overall NMRD effect [see Eqns (1,2) and Table 1]
tHSA shows a b-value of (1.3 ± 0.1)· 107s)2, as
compared to the value of (2.2 ± 0.1)· 107s)2
observed for full-length HSA By assuming that the
b-values obtained by the model-free analysis according
to Eqn (1) of the data shown in Fig 2 are due to
bur-ied water molecules and exchangeable protons, and by
taking into account that the generalized order
para-meter SI is reported to fall in the range 0.5–1 [36], we
should expect that about 51 water molecules are
local-ized within the tertiary structure of tHSA, as
com-pared to 88 water molecules in full-length HSA
Moreover, all of the water molecules appear to be able
to exchange with bulk water in a time longer than the
reorientational correlation time of the protein and
shorter than their own relaxation time [36,37]
There-fore, removal of domain III dramatically affects pro-tein hydration, with a reduction of internal water molecules by a factor of two, independently of the value of the SIparameter (in the range 0.5–1)
Binding of Fe(III)heme to tHSA tHSA contains the complete primary heme-binding site, and shows optical and magnetic spectroscopic properties comparable to those of the full-length pro-tein Heme binds to tHSA, at pH 7.0 and 25C (Fig S1), with Kd= 7.4· 10)8m (i.e K1 in Scheme 1), indicating that the Fe(III)heme affinity for tHSA is slightly higher than that reported for HSA (Kd= 5.0· 10)7m, i.e K5 in Scheme 2 [18]) Heme is known to drive the allosteric transition towards the B-state, thus perturbing molecular contacts between the HSA subdomains that stabilize the N-state [8,38] The affinity constant observed here indicates, there-fore, that the geometry of the Fe(III)heme-binding site
is preserved Moreover, the small, although significant, increase in Fe(III)heme affinity for tHSA could result from the removal of molecular contacts between domains I and III that could hinder the N to B transi-tion in full-length HSA
Figure 3 shows the electronic absorption spectra of Fe(III)heme–tHSA and of full-length Fe(III)heme– HSA For both Fe(III)heme–proteins, the Soret band
Fig 2 NMRD profiles of full-length HSA (filled squares) and tHSA
(open circles), at pH 7.0 and 25 C The protein concentration was
1.0 · 10)3M The continuous lines were obtained by analysis of
the data according to Eqn (1) For details, see text.
Table 1 Parameters obtained from the fitting procedure of NMRD
data in Fig 2 using Eqns (1,2).
K3
Scheme 1 Equilibria for heme and drug binding to tHSA, according
to linked functions [48].
K5
Scheme 2 Equilibria for heme and drug binding to HSA, according
to linked functions [48].
Trang 4is characterized by a maximum at 400 nm, which is
consistent with the high-spin state of the Fe(III) atom
The intensity of the Soret absorption is only slightly
affected on going from pH 7 to pH 11 On the other
hand, a shoulder at 360 nm appears in Fe(III)heme–
tHSA at pH > 9; this spectral change is not observed
in Fe(III)heme–HSA This finding might be accounted
for by significant differences in the B-state of tHSA
with respect to full-length HSA, potentially arising
from the loss of contacts between domains I and III
Relaxometric properties of Fe(III)heme–tHSA
Fe(III)heme–HSA has been widely investigated by 1
H-NMR relaxometry [14,18,19,37] The high value of the
paramagnetic contribution to the paramagnetic
rela-xivity (R1p) of Fe(III)heme–HSA (12.5 mm)1Æs)1 at
0.01 MHz, and 4.0 mm)1Æs)1 at 10 MHz, respectively,
pH 7.0 and 25C) has been formerly ascribed to the occurrence of slowly exchanging water molecules in the surroundings of the paramagnetic Fe(III)heme cen-ter [14,18] Indeed, the high number of incen-ternal wacen-ter molecules calculated above supports this statement The paramagnetic contribution to the solvent water proton relaxation rate observed for Fe(III)heme–HSA
is quite large as compared to oxygen-carrier heme–pro-teins in the ferric form [39–43] Figure 4 shows the NMRD profiles of heme–HSA and heme–tHSA The paramagnetic contribution is dependent on the Larmor frequency, as expected for an S = 5⁄ 2 high-spin sys-tem [44] Owing to the zero field splitting (ZFS) of the
S= 5⁄ 2 manifold, NMRD data cannot be analyzed in terms of the classic Solomon–Bloembergen–Morgan approach [45] In slowly rotating systems, where the electronic relaxation time is shorter than the reorienta-tional correlation time, the ZFS Hamiltonian interacts with the Zeeman Hamiltonian in a time-dependent way, and the electronic relaxation cannot be described simply in terms of electron dipole–dipole interaction Although a rigorous approach would take into account the orientation and the magnitude of the ZFS tensor by numerical methods [46], a set of simplified equations have been proposed to analytically describe the electronic relaxation in S > 1⁄ 2 systems (see Experimental procedures) [47]
By fitting NMRD profiles using Eqns (3–11), a set
of parameters governing the electronic relaxation was obtained (Table 2) It is noteworthy that the exchange lifetime (sM) of the localized water molecules close to the Fe(III)heme does not change significantly The
A
B
Fig 3 Visible region of the electronic absorption spectra of
Fe(III)-heme–tHSA (A) and Fe(III)heme–HSA (B), at 25.0 C The protein
concentration was 1.0 · 10)6M in 1.0 · 10)1M phosphate buffer.
The pH values were changed from 7.0 to 11.0 by using 1.0 M
NaOH (pH 7.0, continuous line; pH 8.0, dotted line; pH 9.0, dash–
dot line; pH 10.0, dash–dot–dot line; pH 11.0, short dash–dot line).
For details, see text.
Fig 4 NMRD profile of 1.0 · 10)3M Fe(III)heme–HSA (filled squares) and of 1.0 · 10)3M Fe(III)heme–tHSA (open circles), at
pH 7.0 and 25 C The continuous lines were obtained by the analy-sis of data according to Eqns (3–11) For details, see text and Experimental procedures.
Trang 5orientation and magnitude of the ZFS tensor, as well
as the correlation time for the static ZFS modulation
(sv), are slightly affected in tHSA with respect to
full-length HSA, reflecting possible rearrangements of the
heme without relevant structural changes; it should be
noted that the reduction of the sv parameter from
3.0· 10)11s in the full-length HSA to 1.5· 10)11s in
the truncated protein reflects the reduction of the
molecular mass and thus the time constant of the static
ZFS modulation On the other hand, the population
of localized water molecules close to Fe(III)heme is
reduced to 53% If it is assumed that, on average, four
water molecules reside at an average distance of 3.3 A˚
from iron ion in Fe(III)heme–HSA, this number is
reduced to 2.1 in Fe(III)heme–tHSA Interestingly, the
fraction of water molecules close to Fe(III)heme
reflects the overall reduction of the number of water
molecules within tHSA (58%) as calculated from the
data in Fig 2 Moreover, the almost coincident
corre-spondence between all the other parameters indicates
that the Fe(III)heme geometry is not significantly
affected by removal of domain III
Drug binding to tHSA and to Fe(III)heme–tHSA
To ascertain whether drug binding affects heme
affin-ity, Fe(III)heme binding to tHSA was investigated in
the presence of ibuprofen and warfarin Analysis of
binding isotherms (Fig S2) according to Eqn (12)
allowed us to obtain Kd= 3.4· 10)6m for
Fe(III)-heme binding to tHSA in the presence of 1.0· 10)4m
ibuprofen, and Kd= 3.0· 10)6m (i.e K3 in
Scheme 1) for Fe(III)heme binding to tHSA in the
presence of 1.0· 10)5m warfarin (Table 2) The
anal-ysis of hyperbolic binding curves (Fig S3) according
to Eqn (12) allowed us to obtain Kd= 1.3· 10)5m
for ibuprofen binding to Fe(III)heme–tHSA, and
Kd= 5.0· 10)6m for warfarin binding to
Fe(III)-heme–tHSA (i.e K4 in Scheme 1) According to linked
functions [48], K2Æ K3= K1Æ K4 Therefore, from the
data reported above, it is possible to obtain the value
of the dissociation equilibrium constants for tHSA– ibuprofen (Kd= 2.8· 10)7m) and tHSA–warfarin (Kd= 1.2· 10)7m) complex formation, respectively (i.e K2in Scheme 1; see Table 4)
For comparison, Fe(III)heme binding to full-length HSA in the presence of drugs was investigated Values
of Kdobtained by data analysis according to Eqn (12) are reported in Table 3 In the presence of ibuprofen (Fig S4), the Kd value for Fe(III)heme–HSA complex formation (Kd= 3.9· 10)6m, i.e K7 in Scheme 2) is similar to the Kdvalue for Fe(III)heme–tHSA complex formation (Kd= 3.4· 10)6m), under the same experi-mental conditions Also in the presence of warfarin, the affinity of Fe(III)heme for HSA (Kd= 1.2·
10)6m, i.e K7 in Scheme 2) is similar to that reported for Fe(III)heme–tHSA complex formation (Kd= 3.0·
10)6m)
Finally, the effect of Fe(III)heme on drug affinity for full-length HSA was taken into account (Fig S5) Ibuprofen binds to Fe(III)heme–HSA with
Kd= 5.4· 10)6m (i.e K8 in Scheme 2) Interestingly, this value is smaller than that obtained for ibuprofen binding to Fe(III)heme–tHSA (Kd= 1.3· 10)5m) under the same experimental conditions, indicating a higher affinity of ibuprofen for full-length HSA As the binding isotherms were obtained by measuring changes in the Soret band of Fe(III)heme, binding of ibuprofen to a site that does not alter the heme envi-ronment would be spectroscopically silent Conversely, the Kd value for warfarin binding to full-length Fe(III)heme–HSA (Kd= 3.1· 10)6m, i.e K8 in Scheme 2) is not significantly different from that obtained for warfarin binding to Fe(III)heme–tHSA (Kd= 5.0· 10)6m) (Table 4) According to linked
Table 2 Parameters obtained from the fitting procedure of NMRD
data in Fig 4 using Eqns (3–11).
Table 4 Values of the equilibrium dissociation constants (K d , M ) for drug binding to tHSA and HSA in the absence and in the presence
of Fe(III)heme, at pH 7.0 and 25 C.
Fe(III)heme–
Fe(III)heme– HSA Warfarin 1.2 · 10)7a 5.0 · 10)6 1.3 · 10)6a 3.1 · 10)6 Ibuprofen 2.8 · 10)7a 1.3 · 10)5 3.9 · 10)6a 5.4 · 10)6
a Calculated according to linked functions (Schemes 1,2).
Table 3 Values of the equilibrium dissociation constants (Kd, M ) for Fe(III)heme binding to tHSA and HSA in the absence and presence
of ibuprofen and warfarin, at pH 7.0 and 25 C.
a
From [18].
Trang 6functions [48], K6Æ K7= K5Æ K8 Therefore, from the
data reported above, it is possible to obtain the value
of the dissociation equilibrium constant for full-length
HSA–ibuprofen (Kd= 3.9· 10)6m) and for
full-length HSA–warfarin (Kd= 1.3· 10)6m, i.e K6 in
Scheme 2) complex formation, respectively (see
Scheme 2 and Table 4)
Conclusion
The data reported here indicate that tHSA is a valuable
model with which to investigate the allosteric properties
of HSA Indeed, by removal of the C-terminal
domai-n III, a domai-number of codomai-ntacts that idomai-nvolve domaidomai-n I (codomai-n-
(con-taining the heme site) and domain II (con(con-taining the
warfarin site) are lost; nevertheless, the allosteric linkage
between the heme and warfarin (i.e Sudlow’s site I) sites
is maintained Moreover, tHSA allows independent
analysis of the linkages between different drug-binding
sites In the case of ibuprofen, for instance, modulation
of Fe(III)heme affinity cannot be attributed to
ibupro-fen binding to either its primary (in domain III)
or secondary (in domain II) binding site in full-length
HSA Indeed, after removal of domain III, ibuprofen
binds to a single site, thus allowing investigation of the
effect of the occupancy of the secondary
ibuprofen-binding site on Fe(III)heme affinity
Finally, it is worth noting that the three ligands
considered here (i.e ibuprofen, warfarin, and heme)
display an increased affinity for tHSA with respect to
HSA If tHSA could fold in a different conformation,
or could not achieve a stable fold, it would be
reason-able to envisage that one or more of the considered
ligands would display reduced or no affinity This
defi-nitely supports the idea that tHSA is a fragment of the
HSA structure with similar folding and similar
confor-mational transitions The analysis of NMRD profiles
of tHSA and Fe(III)heme–tHSA, as well as the
analysis of the optical spectra of Fe(III)heme–tHSA,
are in agreement with this premise
The allosteric properties that make HSA a peculiar
monomeric protein and account for the regulation of
ligand-binding modes by heterotropic interactions are
maintained after the removal of domain III Indeed,
warfarin allosterically inhibits Fe(III)heme binding,
and, in turn, Fe(III)heme allosterically inhibits
warfa-rin binding Moreover, a similar allosteric mechanism
modulates ibuprofen and Fe(III)heme binding to tHSA
that would not occur in the full-length protein
Actu-ally, binding of ibuprofen to the (secondary) tHSA
binding site inhibits Fe(III)heme binding, and, in turn,
Fe(III)heme inhibits ibuprofen binding This finding
explains a former observation that was attributed to
ibuprofen binding to the warfarin site of HSA when the structural description of the ibuprofen-binding mode(s) was not available [9]
In conclusion, a detailed analysis of allosteric mech-anisms that regulate ligand binding to HSA has been made possible by using a simple model protein (tHSA) that maintains the allosteric properties of full-length HSA with a reduced number of binding sites A deep understanding of the functional links between different sites of HSA is essential to avoid critical and unex-pected changes in the pharmacokinetic properties of therapeutic drugs
Experimental procedures
tHSA cloning, expression, and purification
The cDNA sequence of tHSA (corresponding to residues Asp1–Glu382 of HSA) was amplified by PCR from a human liver cDNA library, and cloned into pPICZa-A (In-vitrogen, Carlsbad, CA, USA), downstream of the Saccha-romyces cerevisiae secretion factor, under the control of the AOX1 promoter Primer synthesis and construct sequencing services were provided by MWG Biotech (Ebersberg, Germany) The construct was amplified in Escherichia coli, and subsequently transformed into Pichia pastoris strain GS115 Cells grown in glycerol medium were harvested and resuspended in methanol containing the medium to induce protein synthesis Protein expression in the medium was checked by SDS⁄ PAGE The medium containing the expressed protein was ultrafiltered using a 10 kDa cut-off membrane (Centricon Plus70; Millipore Corporation, Biller-ica, MA, USA), and the concentrated protein was lyophi-lized To remove hydrophobic ligands, the protein was dissolved in water, acidified to pH 3.5 with acetic acid, and treated for 2 h with activated charcoal at room temperature [49] After charcoal removal by centrifugation (20 000 g for
20 min at 2C), the pH was brought to 7.0 with aqueous ammonia The protein concentration was measured accord-ing to Bradford [50], and the solution was then partitioned into aliquots and freeze-dried The integrity of the protein was checked by digestion with trypsin and subsequent MALDI-TOF MS analysis (Reflex III; Bruker Daltonics, Bremen, Germany) All other reagents (Sigma-Aldrich, St Louis, MO, USA) were of the highest purity available, and were used without further purification HSA (Sigma-Aldrich, St Louis, MO, USA) was essentially fatty acid-free, according to the charcoal delipidation protocol [49,51,52], and was used without further purification
Protein and ligand solutions
The Fe(III)heme–tHSA and Fe(III)heme–HSA solutions were prepared by adding the appropriate volume of the
Trang 71.2· 10)2m Fe(III)heme solution, dissolved in
1.0· 10)1m NaOH, to a 1.0· 10)3m protein solution in
0.1 m phosphate buffer (pH 7.0), to a final Fe(III)heme–
protein concentration of 1.0· 10)3m The concentration of
the Fe(III)heme stock solution was checked as
bis-imidazo-late complex in SDS micelles with an extinction coefficient
of 14.5 mm)1cm)1(at 535 nm) [53] The ibuprofen solution
was prepared by dissolving the drug in 1.0· 10)1m
phos-phate buffer, at pH 7.0 and 25.0C The warfarin solution
was prepared by stirring the drug in 1.0· 10)1m
phos-phate buffer at pH 12.0 until it dissolved, and then
adjust-ing the pH to 7.0 with HCl (at 25.0C)
NMRD
NMRD profiles, i.e plots of solvent water proton
relaxa-tion rates as a funcrelaxa-tion of the applied magnetic field, were
measured on a Stelar Spinmaster FFC field cycling
spec-trometer (Stelar, Mede, PV, Italy), operating in a field
range from 2.4· 10)4T to 2.35· 10)1T (corresponding to
proton Larmor frequencies from 0.01 MHz to 10 MHz)
The temperature was set at 25C by using a built-in
tem-perature controller, and directly measured in the probehead
with a mercury thermometer The relaxometer is able to
switch the magnetic field strength in a millisecond
time-scale, and works under complete computer control As a
blank, the measurement of T1 of the buffer solution
(1.0· 10)1m phosphate buffer, pH 7.0) was performed in
the same range of temperatures An absolute uncertainty in
1⁄ T1of about 1%, on average, has been assessed
NMRD profiles of 1.0· 10)3m tHSA and HSA were
analyzed in terms of a model-free approach [35,36],
accord-ing to Eqn (1):
R1ðxÞ ¼ Rwð Þ þ D þ b 1 vT fð Þ 0:2JðxÞ þ 0:8Jð2xÞ½
þ v 0:1Jð0Þ þ 0:3JðxÞ þ 0:6Jð2xÞ½ g ð1Þ
where Rw(T) = 0.9756T· 0.6985 is the relaxation rate of
the blank (i.e of the buffer) solution at any given
tempera-ture T, D is the part of R1(x) that remains in the extreme
motional narrowing regime, b is the mean square
fluctua-tion of the lattice variable coupled to the observed nuclear
spin, and sc is the correlation time for the time-dependent
spin-lattice coupling J(x) is the Lorentzian spectral density
function JðxÞ ¼ sc
1þ ðxscÞ2:
By assuming that the NMRD profile is determined by
water molecules buried within the protein core in
intermedi-ate–fast exchange with bulk water, sc turns out to be the
reorientational correlation time, and the amplitude
parameter A would be related to the number of internal
water molecules (NI) as described hereafter (Eqn 2)
NIS2I ¼b NT
x2 D
ð2Þ
NT is the number of total water molecules (per protein),
and xDis the intramolecular dipole frequency In the case
of hydrogen nuclei, xD= 2.36· 105
radÆs)1 SI is the mean-square generalized order parameter for the internal water molecules, and cannot be > 1 [39]
NMRD profiles of 1.0· 10)3m Fe(III)heme–tHSA and Fe(III)heme–HSA were obtained by subtracting from the measured relaxation rate the relaxation rate of the corre-sponding apoprotein (i.e tHSA and HSA) at the same frequency Profiles were analyzed in terms of Eqns (3–11) [47]:
R1p¼ Nq
T1m¼ Rð 1zþ R1xÞ1 ð4Þ
R1z¼35 3
K
r6
U1ð Þhz sSz
1þ x2
Is2 Sx
ð5Þ
R1x¼2 3
K
r6
U2ð Þhz 10sSx 1þ16c2D2s2
Sx
þ 16sSx 1þ4c2D2s2
Sx
þ 9sSx 1þx2
Is2 Sx
ð6Þ
K ¼15 2
l0 4p
h 2p
2
c2
sc2
U1ð Þ ¼hz 1þ P2ðcos hzÞ
U2ð Þ ¼hz 2 P2ðsin hzÞ
s1Sz ¼ 2
35½4SðS þ 1Þ 3D
2
1þ 4c2D2s2
v
þ 80sm
1þ 16c2D2s2
v
þ 160sv
1þ 36c2D2s2
v
ð10Þ
s1Sx ¼2
35½4SðS þ 1Þ 3D
2
168svþ 152sv
1þ 4c2D2s2
v
þ 200sv 1þ 16c2D2s2
v
þ 40sv 1þ 36c2D2s2
v
ð11Þ
where N is the molar concentration of Fe(III)heme, q is the number of water molecules coordinated to the metal ion,
r is the average distance between the metal ion and the protons of the water molecules, sMis their mean residence lifetime, xI is the proton Larmor frequency, P2(x) is the second-order Legendre polynomial, sv is the correlation time of the modulation of the transient ZFS, D is the aver-age energy of the electron–ZFS coupling, D is the energy separation of ZFS levels, h is the orientation of the ZFS tensor in the molecular frame with respect to the laboratory frame, c is the speed of light, l0 is the permeability of vacuum, h is the Planck constant, S is the electron spin quantum number, and cS and cI are the electron and the proton nuclear magnetogyric ratios, respectively
Trang 8Optical binding studies
Fe(III)heme binding to HSA and tHSA, in the absence
and presence of 1.0· 10)4m ibuprofen and 1.0· 10)5m
warfarin, was investigated spectrophotometrically using an
optical cell with 1.0 cm path length on a Cary 50 Bio
spectrophotometer (Varian Inc., Palo Alto, CA, USA) In
experiments carried out at different Fe(III)heme
concen-trations, a small amount of the 1.0· 10)3m HSA or
tHSA solution was diluted in the optical cell in
1.0· 10)1m phosphate buffer and 10% dimethylsulfoxide
(pH 7.0), to a final protein concentration of 1.0· 10)6m
Then, small amounts of Fe(III)heme (1.2· 10)2m) were
added to the protein solution, and the absorbance spectra
were recorded after incubation for few minutes, after
each addition In experiments carried out at different
drug concentrations, a small amount of Fe(III)heme
(1.2· 10)2m) and of HSA solution (about 1.0· 10)3m)
was diluted in the optical cell in 1.0· 10)1m phosphate
buffer and 10% dimethylsulfoxide (pH 7.0), to a final
Fe(III)heme–HSA or Fe(III)heme–tHSA concentration of
1.0· 10)6m Then, small aliquots of 1.0· 10)3m
ibuprofen or 2.0· 10)2m warfarin were added to the
Fe(III)heme–protein solution, and the absorbance spectra
were recorded after incubation for a few minutes after
each addition Binding isotherms were analyzed by
plotting the absorbance change as a function of the
ligand concentration Data were analyzed according to
Eqn (12):
where DA is the difference in the Soret band (400 nm)
absorbance, DAmax is the absorbance difference at
saturating ligand concentration, Kd is the dissociation
equilibrium constant for ligand–protein complex
formation, [Lt] is the total concentration of the variable
ligand [Fe(III)heme, warfarin, or ibuprofen], [Pt] is the
total concentration of the protein [(t)HSA, Fe(III)heme–
(t)HSA, warfarin–(t)HSA, or ibuprofen–(t)HSA], and N
is the number of equivalent binding sites (N = 1 for
both tHSA and HSA for each of the three ligands
considered)
Acknowledgements
We gratefully acknowledge S Aime and S Baroni for
helpful discussions
References
1 Fasano M, Curry S, Terreno E, Galliano M, Fanali G, Narciso P, Notari S & Ascenzi P (2005) The extraordi-nary ligand binding properties of human serum albu-min IUBMB Life 57, 787–796
2 Ascenzi P, Bocedi A, Notari S, Fanali G, Fesce
R & Fasano M (2006) Allosteric modulation of drug binding to human serum albumin Mini Rev Med Chem
6, 483–489
3 Peters T Jr (ed.) (1996) All about Albumin: Biochemistry, Genetics and Medical Applications Academic Press, San Diego; London
4 Curry S, Mandelkov H, Brick P & Franks N (1998) Crystal structure of human serum albumin complexed with fatty acid reveals an asymmetric distribution of binding sites Nat Struct Biol 5, 827–835
5 Curry S (2002) Beyond expansion: structural studies on the transport roles of human serum albumin Vox Sang
83, 315–319
6 Sudlow G, Birkett DJ & Wade DN (1975) The charac-terization of two specific drug binding sites on human serum albumin Mol Pharmacol 11, 824–832
7 Dockal M, Chang M, Carter DC & Ru¨ker F (2000) Five recombinant fragments of human serum albumin Tools for the characterization of the warfarin binding site Protein Sci 9, 1455–1465
8 Petitpas I, Bhattacharya AA, Twine S, East M & Curry
S (2001) Crystal structure analysis of warfarin binding
to human serum albumin: anatomy of drug site I J Biol Chem 276, 22804–22809
9 Baroni S, Mattu M, Vannini A, Cipollone R, Aime S, Ascenzi P & Fasano M (2001) Effect of ibuprofen and warfarin on the allosteric properties of haem–human serum albumin A spectroscopic study Eur J Biochem
268, 6214–6220
10 Hage DS, Noctor TA & Wainer IW (1995) Character-ization of the protein binding of chiral drugs by high-performance affinity chromatography Interactions of R- and S-ibuprofen with human serum albumin
J Chromatogr A 693, 23–32
11 Chen J, Fitos I & Hage DS (2006) Chromatographic analysis of allosteric effects between ibuprofen and ben-zodiazepines on human serum albumin Chirality 18, 24–36
DA¼
DAmax Kd1 L½ þ N Pt ½ Kt 1
d þ 1
ffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
Kd1 L½ þ N Pt ½ Kt 1
d þ 1
4K2
d L½ N Pt ½ t q
2K1
Trang 912 Ghuman J, Zunszain PA, Petitpas I, Bhattacharya AA,
Otagiri M & Curry S (2005) Structural basis of the
drug-binding specificity of human serum albumin
J Mol Biol 353, 38–52
13 Sugio S, Kashima A, Mochizuki S, Noda M &
Kobay-ashi K (1999) Crystal structure of human serum
albu-min at 2.5 A˚ resolution Protein Eng 12, 439–446
14 Fasano M, Baroni S, Vannini A, Ascenzi P & Aime S
(2001) Relaxometric characterization of human
hemal-bumin J Biol Inorg Chem 6, 650–658
15 Mattu M, Vannini A, Coletta M, Fasano M & Ascenzi
P (2001) Effect of bezafibrate and clofibrate on the
heme-iron geometry of ferrous nitrosylated heme–
human serum albumin: an EPR study J Inorg Biochem
84, 293–296
16 Fasano M, Mattu M, Coletta M & Ascenzi P (2002)
The heme-iron geometry of ferrous nitrosylated heme–
serum lipoproteins, hemopexin, and albumin: a
compar-ative EPR study J Inorg Biochem 91, 487–490
17 Monzani E, Curto M, Galliano M, Minchiotti L, Aime
S, Baroni S, Fasano M, Amoresano A, Salzano AM,
Pucci P et al (2002) Binding and relaxometric
proper-ties of heme complexes with cyanogen bromide
fragments of human serum albumin Biophys J 83,
2248–2258
18 Fanali G, Bocedi A, Ascenzi P & Fasano M (2007)
Modulation of heme and myristate binding to human
serum albumin by anti-HIV drugs An optical and
NMR spectroscopic study FEBS J 274, 4491–4502
19 Fanali G, Ascenzi P & Fasano M (2007) Effect of
prototypic drugs ibuprofen and warfarin on global
chaotropic unfolding of human serum heme–albumin: a
fast-field-cycling1H-NMR relaxometric study Biophys
Chem 129, 29–35
20 Nicoletti FP, Howes BD, Fittipaldi M, Fanali G, Fasano
M, Ascenzi P & Smulevich G (2008) Ibuprofen induces
an allosteric conformational transition in the heme
com-plex of human serum albumin with significant effects on
heme ligation J Am Chem Soc 130, 11677–11688
21 Wardell M, Wang Z, Ho JX, Robert J, Ru¨ker F, Ruble
J & Carter DC (2002) The atomic structure of human
methemalbumin at 1.9 A˚ Biochem Biophys Res
Com-mun 291, 813–819
22 Zunszain PA, Ghuman J, Komatsu T, Tsuchida E &
Curry S (2003) Crystal structural analysis of human
serum albumin complexed with hemin and fatty acid
BMC Struct Biol 3, 6, doi:10.1186/1472-6807-3-6
23 Simard JR, Zunszain PA, Hamilton JA & Curry S
(2006) Location of high and low affinity fatty acid
binding sites on human serum albumin revealed by
NMR drug-competition analysis J Mol Biol 361,
336–351
24 Fasano M, Fanali G, Leboffe L & Ascenzi P (2007)
Heme binding to albuminoid proteins is the result of
recent evolution IUBMB Life 59, 436–440
25 Chuang VTG & Otagiri M (2002) How do fatty acids cause allosteric binding of drugs to human serum albu-min? Pharm Res 19, 1458–1464
26 Fitos I, Visy J & Kardos J (2002) Stereoselective kinet-ics of warfarin binding to human serum albumin: effect
of an allosteric interaction Chirality 14, 442–448
27 Chen J, Ohnmacht C & Hage DS (2004) Studies of phenytoin binding to human serum albumin by high-performance affinity chromatography J Chromatogr
B Analyt Technol Biomed Life Sci 809, 137–145
28 Ascenzi P, Bocedi A, Bolli A, Fasano M, Notari S & Polticelli F (2005) Allosteric modulation of monomeric proteins Biochem Mol Biol Educ 33, 169–176
29 Kim HS & Hage DS (2005) Chromatographic analysis
of carbamazepine binding to human serum albumin
J Chromatogr B Analyt Technol Biomed Life Sci 816, 57–66
30 Ascenzi P & Fasano M (2007) Abacavir modulates per-oxynitrite-mediated oxidation of ferrous nitrosylated human serum heme–albumin Biochem Biophys Res Commun 353, 469–474
31 Wilting J, van der Giesen WF, Janssen LH, Weideman
MM, Otagiri M & Perrin JH (1980) The effect of albu-min conformation on the binding of warfarin to human serum albumin The dependence of the binding of war-farin to human serum albumin on the hydrogen, calcium, and chloride ion concentrations as studied by circular dichroism, fluorescence, and equilibrium dialy-sis J Biol Chem 255, 3032–3037
32 Yamasaki K, Maruyama T, Yoshimoto K, Tsutsumi Y, Narazaki R, Fukuhara A, Kragh-Hansen U & Otagiri
M (1999) Interactive binding to the two principal ligand binding sites of human serum albumin: effect of the neutral-to-base transition Biochim Biophys Acta 1432, 313–323
33 Sakai H, Masada Y, Horinouchi H, Yamamoto M, Ikeda E, Takeoka S, Kobayashi K & Tsuchida E (2004) Hemoglobin-vesicles suspended in recombinant human serum albumin for resuscitation from hemorrhagic shock in anesthetized rats Crit Care Med 32, 539–545
34 Dockal M, Carter DC & Ru¨ker F (1999) The three recombinant domains of human serum albumin Struc-tural characterization and ligand binding properties
J Biol Chem 274, 29303–29310
35 Bertini I, Fragai M, Luchinat C & Parigi G (2000)
1H NMRD profiles of diamagnetic proteins: a model-free analysis Magn Res Chem 38, 543–550
36 Halle B, Denisov VP & Venu K (1999) Multinuclear relaxation dispersion studies of protein hydration In Biological Magnetic Resonance: Structure Computation and Dynamics in Protein NMR(Krishna NR & Berliner
LJ, eds), 17, 419–484 Kluwer Academic⁄ Plenum Publishers, New York
37 Fasano M, Orsale M, Melino S, Nicolai E, Forlani F, Rosato N, Cicero D, Pagani S & Paci M (2003) Surface
Trang 10changes and role of buried water molecules during the
sulfane sulfur transfer in Rhodanese from Azotobacter
vinelandii: a fluorescence quenching,15N NMR
relaxa-tion, and nuclear magnetic relaxation dispersion
spectroscopic study Biochemistry 42, 8550–8557
38 Fanali G, Fesce R, Agrati C, Ascenzi P & Fasano M
(2005) Allosteric modulation of myristate and
Mn(III)-heme binding to human serum albumin – optical and
NMR spectroscopy characterization FEBS J 272,
4672–4683
39 Fabry ME & Eisenstadt M (1974) The mechanism of
water proton nuclear magnetic resonance relaxation in
the presence of mammalian and Aplysia metmyoglobin
fluoride J Biol Chem 249, 2915–2919
40 Giacometti GM, Ascenzi P, Brunori M, Rigatti G,
Gia-cometti G & Bolognesi M (1981) Absence of water at
the sixth co-ordination site in ferric Aplysia myoglobin
J Mol Biol 151, 315–319
41 Modi S, Behere DV, Mitra S & Bendall DS (1991)
Coordination geometry of haem in cyanogen bromide
modified myoglobin and its effect on the formation
of compound I J Chem Soc Chem Commun 1991, 830–
831, doi:10.1039/C39910000830
42 Aime S, Ascenzi P, Fasano M & Paoletti S (1993)
NMR relaxometric studies of water accessibility to
haem cavity in horse heart and sperm whale myoglobin
Magn Reson Chem 31, S85–S89
43 Aime S, Fasano M, Paoletti S, Cutruzzola` F, Desideri
A, Bolognesi M, Rizzi M & Ascenzi P (1996) Structural
determinants of fluoride and formate binding to
hemo-globin and myohemo-globin: crystallographic and 1H-NMR
relaxometric study Biophys J 70, 482–488
44 Banci L, Bertini I & Luchinat C (1991) Electron and
Nuclear Relaxation VCH, Weinheim
45 Schaefle N & Sharp R (2004) Electron spin relaxation
due to reorientation of a permanent zero field splitting
tensor J Chem Phys 121, 5387–5394
46 Belorizky E, Fries PH, Helm L, Kowalewski J, Kruk
D, Sharp RR & Westlund PO (2008) Comparison of
different methods for calculating the paramagnetic
relaxation enhancement of nuclear spins as a function
of the magnetic field J Chem Phys 128, 052315,
doi: 10.1063/1.2833957
47 Sharp RR (2002) Closed-form expressions for
level-averaged electron spin relaxation times outside the
Zeeman Limit: application to paramagnetic NMR relaxation J Magn Res 154, 269–279
48 Wyman J (1964) Linked functions and reciprocal effects
in hemoglobin: a second look Adv Protein Chem 19, 223–286
49 Cabrera-Crespo J, Goncalves VM, Martins EA, Grellet
S, Lopes AP & Raw I (2000) Albumin purification from human placenta Biotechnol Appl Biochem 31, 101–106
50 Bradford MM (1976) A rapid and sensitive method for the quantitation of microgram quantities of protein utilizing the principle of protein-dye binding Anal Biochem 72, 248–254
51 Sogami M & Foster JF (1968) Isomerization reactions of charcoal-defatted bovine plasma albumin The N–F tran-sition and acid expansion Biochemistry 7, 2172–2182
52 Chen RF (1967) Removal of fatty acids from serum albumin by charcoal treatment J Biol Chem 242, 173– 181
53 Boffi A, Das TK, Della Longa S, Spagnuolo C & Rous-seau DL (1999) Pentacoordinate hemin derivatives in sodium dodecyl sulfate micelles: model systems for the assignment of the fifth ligand in ferric heme proteins Biophys J 77, 1143–1149
Supporting information
The following supplementary material is available: Fig S1 Binding isotherm for Fe(III)heme–tHSA com-plex formation, at pH 7.0 and 25C
Fig S2 Fe(III)heme binding to tHSA, at pH 7.0 and
25C
Fig S3 Drug binding to Fe(III)heme–tHSA, at
pH 7.0 and 25C
Fig S4 Fe(III)heme binding to HSA in the presence
of drugs, at pH 7.0 and 25C
Fig S5 Drug binding to Fe(III)heme–HSA, at pH 7.0 and 25C
This supplementary material can be found in the online version of this article
Please note: Wiley-Blackwell is not responsible for the content or functionality of any supplementary materials supplied by the authors Any queries (other than missing material) should be directed to the corre-sponding author for the article