A structural resemblance was observed between the native protein and the structurally perturbed state which resulted after heat treatment at 110C.. However, inspection of the resolved pa
Trang 1Sotirios Koutsopoulos1, John van der Oost2 and Willem Norde1,3
1 Laboratory of Physical Chemistry and Colloid Science, Wageningen University, the Netherlands
2 Laboratory of Microbiology, Wageningen University, the Netherlands
3 Department of Biomedical Engineering, University Medical Center Groningen, the Netherlands
Hyperthermophilic microorganisms predominantly
belong to the Archaea, the third phylogenetic domain
of life [1] They flourish in environments of extreme
temperatures even higher than 100C, which until
recently were considered as incompatible with life No
multicellular organisms have been found to tolerate
temperatures above 60C and no unicellular eukarya
have been discovered to withstand long-term exposure
to temperatures higher than 70 C Pyrococcus
furio-sus is an anaerobic hyperthermophile which was
dis-covered in geothermally heated marine sediments at
100C [2] It is a very efficient consumer of the
organic material found on the sea floor such as
pro-teins, peptides and sugar mixtures (e.g maltose,
cello-biose, oligosaccharides and starch), which are
fermented and used as carbon source P furiosus has
a large collection of hyperthermostable enzymes
which may be used in important applications in
biotechnology One of them, the extracellular endo-b-1,3-glucanase (LamA), has been isolated and charac-terized [3] LamA hydrolyzes 1,3-b-glycosyl bonds of polysaccharides such as laminarin The temperature of maximum activity is 104C and the optimal pH 6.5 LamA is practically inactive at room temperature and shows detectable activity only above 30C [4]
The intrinsic fluorescence from LamA’s tryptophans can be used to study its structural characteristics and identify conformational states upon heat and chemical treatment [5] The fluorescence emission spectrum of proteins depends on the microenvironment of the fluorescent amino acids Fluorescence spectroscopy is a useful technique for studying partially folded or unfol-ded proteins; NMR and X-ray crystallography are much less practical due to the structural heterogeneity and mobility of the polypeptide chain In the steady-state fluorescence measurements the sample is
Keywords
circular dichroism; hyperthermostable
protein; steady-state fluorescence;
time-resolved fluorescence and anisotropy
Correspondence
S Koutsopoulos, Center for Biomedical
Engineering, Massachusetts Institute of
Technology, NE47-Room 307, 500
Technology Square, Main Street,
Cambridge, MA 02139-4307, USA
Fax: +1 617 258 5239
Tel: +1 617 324 7612
E-mail: sotiris@mit.edu
(Received 15 July 2005, accepted 30 August
2005)
doi:10.1111/j.1742-4658.2005.04941.x
The structural features of the hyperthermophilic endo-b-1,3-glucanase from Pyrococcus furiosus were studied using circular dichroism, steady-state and time-resolved fluorescence spectroscopy and anisotropy Upon heat and chemical treatment the folded and denatured states of the protein were characterized by distinguishable spectral profiles that identified a number
of conformational states The fluorescence methods showed that the spec-tral differences arose from changes in the local environment around specific tryptophan residues in the native, partially folded, partially unfolded and completely unfolded state A structural resemblance was observed between the native protein and the structurally perturbed state which resulted after heat treatment at 110C The enzyme underwent disruption of the native secondary and tertiary structure only after incubation at biologically extre-mely high temperatures (i.e 150C), whilst in the presence of 8 m of guani-dine hydrochloride the protein was partially unfolded
Abbreviations
ANS, 8-anilino naphthalene-1-sulfonic acid; CD, circular dichroism; GdnHCl, guanidine hydrochloride; LamA, endo-b-1,3-glucanase.
Trang 2constantly illuminated and the emission is recorded.
Time-resolved measurements are performed with
expo-sure of the sample to a picosecond light pulse and
recording of the intensity decay in the nanosecond
timescale [6] The fluorescence and anisotropy decays
contain information on the shape, rigidity,
compact-ness, fluorophore dynamics and rotational motion of
the protein [6,7] Even in the absence of structural
data, valuable information about the local and global
dynamics of LamA can be inferred from inspection of
the fluorescence decays alone
In this study, the structural characteristics of the
hyperthermostable LamA are investigated at extreme
temperatures and high concentrations of guanidine
hydrochloride (GdnHCl) The spectroscopic analysis
will enable us to characterize the thermally and
chem-ically denatured states of LamA Using a combination
of circular dichroism, steady-state and time-resolved
spectroscopy and anisotropy we will show that it is
possible to observe conformations of partially
struc-tured, partially unfolded and completely unfolded
states, depending on the treatment
Results
The hyperthermophilic LamA is a single-domain protein
with a molar mass of 30 085 Da Experimental data
from mass spectroscopy (MALDI TOF) and size
exclu-sion chromatography showed that LamA in solution is
a monomer LamA contains 11 tryptophans
homogen-ously distributed over the amino acid sequence (Fig 1)
For the graphical representation a molecular simulation,
software was utilized [8] assuming structural similarity
of LamA with a homologous 1,3-1,4-b-glucanase from
Bacillus licheniformis and with a j-carrageenase
frag-ment from Pseudoalteromonas carrageenovora whose
crystal structures are known (PDB entries 1GBG and
1DYP, respectively) [9,10] According to the model, the
shape of LamA is globular-ellipsoid with calculated
dimensions of 4.6 nm· 3.2 nm · 3.4 nm For the
selec-tion of the best model preliminary analysis of the NMR
solution structure of LamA as well as spectroscopic data
from this work were taken into consideration
Investiga-tion of proteins with multiple tryptophans results in
emission spectra that represent the contribution from all
emitting groups Nevertheless, valuable information can
be obtained from analyses of the conformational states
of LamA upon heat treatment and in the presence of
GdnHCl At the experimental conditions employed in
this work LamA shows a calorimetric transition at
109C which represents denaturation [11] and
main-tains its structural integrity at high concentrations of
GdnHCl up to 5.5 m
Circular dichroism (CD) The secondary and tertiary structural features of LamA were studied by far- and near-UV CD, respect-ively As may be seen in Fig 2 (top, curve a), the
far-UV CD spectrum of native LamA exhibits a broad negative peak at 217 nm and a positive absorption dif-ference band from 207 nm This spectral profile is characteristic of proteins predominantly consisting of b-structures The spectral analysis revealed that native LamA consists of b-sheets and turns up to 96% (Table 1) Heating the protein solution up to 98C followed by cooling resulted in restoring the spectral ellipticity (spectrum coincided with curve a in the top panel of Fig 2) However, heating at and above the
Fig 1 Graphic display of the structure of LamA using molecular modeling The enzymatic cleft is located on the top of the structural representation Secondary structural elements (A) and the position (B) of the tryptophans in the three-dimensional structure (graphs were generated with Swiss PDB Viewer).
Trang 3denaturation temperature (e.g 110C) did not result
in recovering the spectral features of the native protein
(Fig 2; top, curve b) Notably, heating at such high
temperatures did not unfold the hyperthermostable
protein The features of the native state could still be observed in the denatured sample, illustrating the per-sistence of a stable network of b-structures up to
87% Monitoring the ellipticity at 220 nm showed the beginning of the thermal transition which indicated that at 110C (i.e., just above the denaturation point
of the protein) residual secondary structure was still present (Fig 2, inset) The CD spectrum of LamA at
110 C closely resembled the one recorded for the same sample after cooling to room temperature Heat incubation at 150C for 30 min resulted in collapsed secondary structure and the polypeptide chain appeared to be unordered (Fig 2; top, curve c) Severe changes in the secondary structure were also observed in the presence of 8 m GdnHCl but the effect could not be quantified (Fig 2; top, curve d) This finding is in contrast to a previous study where it was reported that the presence of 7.9 m GdnHCl did not alter LamA’s secondary structural characteristics [12]
In the near-UV region the differences between the native and the thermally denatured states were more noticable The CD spectrum of native LamA shows two minima in ellipticity at 295 nm and 265 nm The bands arose from the aromatic residues fixed in
an asymmetric environment The CD spectrum of de-natured LamA after heating at 110C resembled the one of native LamA but the intensities of the bands were lower After heat incubation at 150C the spec-trum of LamA had very little and no ellipticity at
295 nm and 262 nm, respectively (Fig 2; bottom, curve c), suggesting disruption of the tertiary structure
In the presence of 8 m GdnHCl the near-UV CD spec-tral profile of LamA showed decreased ellipticity of the bands around 295 and 262 nm and increased ellip-ticity of the positive band around 285 nm (Fig 2; bottom, curve d) These changes, although significant, strongly suggest that even at 8 m GdnHCl the protein did not completely unfold These results are in agree-ment with data reported by Chiaraluce et al [12]
Steady-state fluorescence spectroscopy The fluorescence emission spectra of LamA recorded after excitation at 300 nm are typical for a multitryp-tophan protein [6] The native protein shows a maxi-mum at 335 nm (Fig 3; curve a) After heating of the protein solution to 110C the maximum intensity shif-ted to 344 nm This indicates partial exposure of tryp-tophan(s) to water, possibly due to a structural distortion Incubation at 150C shifted the emission maximum to 356 nm suggesting significant exposure of tryptophans and possibly collapsed tertiary structure Interesting features were also revealed from the
-15
-10
-5
0
5
10
15
20
25
Wavelength (nm)
[ θ
1- )
(a) (b)
(c) (d)
-0.4
-0.2
0.0
0.2
0.4
(b)
(d)
(c)
(a)
(c)
(b) (d)
(a)
1- )
Wavelength (nm)
-15 -10 -5 0 5
Temperature (oC) [ θ
1- )
phosphate buffer at pH 7.0 in the far-UV (top) and near-UV (bottom)
region of the spectrum Lines represent: (a) LamA in the native
the molar ellipticity at 220 nm.
phosphate at pH 7.0 in the native state, after heat treatment and in
Sample
a-helix
(%)
b-sheet (%)
b-turn (%)
Unordered (%)
Trang 4respective fluorescence intensities Heating LamA to
110C resulted in decreased emission The effect of
thermal treatment was more pronounced after
incuba-tion at 150C and subsequent recording of the
fluores-cence emission at 20C (i.e., the intensity decreased
threefold as compared to that of native LamA) The
emission maximum of LamA in 8 m GdnHCl was
observed at 350 nm with two-fold increased intensity
(Fig 3)
8-Anilino naphthalene-1-sulfonic acid (ANS)
fluorescence spectroscopy measurements
Coherence and integrity of the external surface of
LamA upon thermal and chemical treatment were
tes-ted by measuring the exposure of the hydrophobic
groups to the solvent The fluorescence intensity of
ANS is quenched in aqueous solution, but in contact
with nonpolar groups a striking emission enhancement
is observed [13,14] Depending on the treatment, the
interaction of LamA with ANS resulted in notable
dif-ferences in the fluorescence emission of the probe
(Fig 4) Heating LamA at 110C resulted in 12-fold
increased intensity relative to that of the native state
After incubation at 150C the intensity was similar to
that of native LamA but the ANS emission maximum
was clearly blue-shifted to 460 nm (Fig 4; curve c)
which suggests increased exposure of hydrophobic
groups In the presence of 8 m GdnHCl the ANS
fluor-escence could not be measured, probably due to the
interaction of ANS with the denaturant
Time-resolved fluorescence decay
In an attempt to understand the origin of the differ-ences observed in the steady-state fluorescence spectra,
we inspected the time-resolved profiles The decays were best fitted by five components according to Eqn (4) (Experimental procedures), except in the case
of LamA in 8 m GdnHCl where four exponents were sufficient The lifetimes (s) and their fractional contri-butions (a) associated with the decays are summarized
in Table 2 Heat and chemical treatment of LamA (Fig 5; curves b–d) resulted in fluorescence decays that relaxed at longer lifetimes as compared to that of the native state (Fig 5; curve a) This can also be seen in Table 2, from the increased contribution (ai) of the longest lifetimes (si) on the average fluorescence life-times, <s> LamA thermally treated at 110C has a dynamic fluorescence profile that clearly differs from that of the native protein The differences are striking
as compared to the information obtained from the steady-state spectra (Fig 3; curves a and b) Compar-ison of the decays justifies the dynamic diversity of the tryptophans’ local microenvironment owing to conformational changes Heat treatment at 110C and incubation at 150C resulted in similar decay profiles However, inspection of the resolved parameters shows that after heat treatment at 110C, the short fluores-cence lifetimes, s1–s3, are shorter relative to those found for LamA after incubation at 150C The pic-ture is reversed at longer lifetimes (Table 2) In the
0 20 40 60 80 100 120
Wavelength (nm)
(b)
(c)
(d)
(a)
Fig 4 Binding of ANS to LamA before and after heat and chemical
fluor-escence of LamA (a) in the native state, (b) thermally denatured at
GdnHCl.
0
50
100
150
200
250
300
350
Wavenumber (nm)
(a)
(b)
(c)
(d)
Fig 3 Normalized steady-state fluorescence emission spectra of
LamA in sodium phosphate buffer at pH 7.0 Curve (a) refers to the
wavelength was 300 nm.
Trang 5presence of 8 m GdnHCl the decay differs from that of native and heat-treated LamA In this case, the data analysis showed that the shortest and the longest life-times observed in the other samples could not be resolved Instead, the major contribution to the decay arises from tryptophans relaxing at medium and relat-ively long lifetimes
Time-resolved anisotropy decay Two exponential terms were required to describe the anisotropy decays of LamA according to Eqn (6) The fitting parameters are summarized in Table 3 The fluorescence is mainly depolarized by the rapid local motion of the tryptophans and by the overall rotation
of the entire protein The diversity of the anisotropy decays observed for each sample (Fig 6) suggests a different depopulation mechanism of the excited state depending on the protein conformation The aniso-tropy of native LamA decays slower relative to that after heat and chemical treatment Data analysis revealed two rotational correlation times at /1¼
260 ps and /2¼ 18.9 ns with amplitudes b1¼ 0.038 and b2 ¼ 0.122, respectively The shortest correlation time is associated with the rapid internal flexibility of a population of indole rings, which depends on the microenvironment that the tryptophans reside in, in the protein The longer component is relevant to the rotational diffusion of the protein from which the hydrodynamic size may be calculated using the Einstein–Stokes equation (u ¼ 4pR3
hg=3kT; where g is the viscosity of the medium, k is the Boltzmann con-stant and T is the absolute temperature) The hydro-dynamic radius, Rh, of native LamA was found to be 2.63 nm This value is in good agreement with the pro-tein size of the model, especially if the size of the hydration layer surrounding the protein in solution is taken into account After thermal denaturation at
110 C, the anisotropy decay was found to be consid-erably different from that observed for the native state (Fig 6; curves a and b) This is also shown in the short correlation time resolved at /1¼ 434 ps which is longer than that observed in native LamA but which has a larger amplitude The longest rotational correla-tion time is slightly longer than that of the native state but the difference in the calculated hydrodynamic radii does not document size expansion (Table 3) After incubation at 150C and in the presence of 8 m GdnHCl, the anisotropy decayed very fast and calcula-tions on the size of the protein could not be implemen-ted In the case of incubation at 150C the longest correlation time, which was observed in the native state and in LamA after heating at 110C, could not
a1
s1
a2
s 2
a3
s 3
a4
s 4
a5
s5
Trang 6be resolved Instead, a medium-lived component was
found at 3.8 ns It should be noted that the long
rota-tional correlation time observed for LamA in the
presence of 8 m GdnHCl should be corrected by a
factor 2.3 when compared to the respective lifetimes of LamA in guanidine-free solutions [15] This is due to the difference between the viscosity of the solution in the presence and in the absence of 8 m GdnHCl, i.e.,
at 67% confidence intervals ND, not determined.
LamA in solution 0.038 (0.027–0.048) 0.26 (0.24–0.28) 0.122 (0.111–0.132) 18.90 (17.97–19.84) 2.63 (2.59–2.68) 0.16 23.5 (21.2–24.9)
GdnHCl
1 10 100 1000 10000 100000 1000000
Time (ns)
(a)
(b) (c) (d)
-0,04 0 0,04
-0,04 0 0,04
-0,04 0 0,04
-0,04 0 0,04
-0,3 0 0,3
-0,3 0 0,3
-0,3 0 0,3
-0,3 0 0,3
(a)
(c) (b)
(d)
(a)
(c) (b)
(d)
Fig 5 Time-resolved fluorescence decay of
295 nm The y axis is in a logarithmic scale.
The lines represent the fluorescence decay
of (a) LamA in the native state, (b) heated at
Trang 72.39 cP and 1.02 cP at 20C, respectively [16] Short
life-times are not affected by the viscosity of the medium
The fundamental anisotropy, ro, representing the
total anisotropy in the absence of rotation (at t¼ 0),
is equal to the sum of the amplitudes, bi, of the
fluoro-phores For excitation at 295 nm the theoretical
time-zero anisotropy is about 0.3 [6] This is higher than the
values obtained for the tryptophans in LamA
Depend-ing on the protein conformation and the freedom of
the tryptophans to rotate in the protein matrix, the ro
may be reduced as a result of subpicosecond motions
that are too fast to be detected [17,18], noncollinearity
of the absorption and emission dipoles [6,18,19], and
intertryptophan energy migration [20,21]
The rotation angle, h, of the tryptophans attached in the protein backbone may be calculated from the amplitude b1of the fast motion [7]:
1bfast
ro
¼3 cos
2h 1
The average cones of rotation of the tryptophans in LamA (Table 3) increase from 23.5 in the native state
to 32.4 in the thermally denatured state, to 39.7 in the unfolded state after incubation at 150C, to 49.0
in the perturbed conformation in the presence of 8 m GdnHCl The increase of the rotational freedom in the heat and chemically treated LamA illustrates the fast anisotropy decays observed in Fig 6
0,00
0,05
0,10
0,15
0,20
Time (ns)
(a)
(b)
(c) (d)
-10
0
10
-10
0
10
-10
0
10
-10
0
10
-0,2 0 0,2
-0,2 0 0,2
-0,2 0 0,2
-0,2 0 0,2
(a)
(c) (b)
(d)
(a)
(c) (b)
(d)
Fig 6 Time-resolved anisotropy of
represent (a) native LamA, (b) LamA heat
GdnHCl The lines represent fitting of the anisotropy exponential decay with two components as shown in Table 3.
Trang 8The hyperthermostable LamA shows a heat
denatura-tion transidenatura-tion at 109C and only a partly unfolded
structure at 7.9 m GdnHCl [11,12] In this study, the
structural characteristics of LamA were thoroughly
investigated upon thermal and chemical treatment The
spectroscopic data suggested different conformations
depending on the temperature of the treatment It was
shown that after cooling to room temperature, the
thermally denatured LamA did not refold to the native
conformation but to a compact form with defined
structure that is different from that of the native and of
the chemically denatured states Such a conformation
resembles the features of a molten globule exhibiting
native-like secondary structure but different tertiary
structure LamA’s irreversible denaturation is
con-firmed by calorimetric experiments (S Koutsopoulos,
J van der Oost & W Norde, unpublished data) Both
the secondary and tertiary structures irreversibly
col-lapsed only after prolonged heating at 150C The
interaction of LamA with GdnHCl solutions did not
show significant changes in the spectroscopic
character-istics of the protein up to 5.5 m GdnHCl Severe
changes in LamA’s secondary and tertiary structure
were observed in the presence of 8 m GdnHCl
Inspection of the far-UV CD spectra showed minor
changes in the secondary structure of LamA upon heat
denaturation at 110C while significant changes were
observed upon incubation at 150C and in the
pres-ence of 8 m GdnHCl Moreover, the near-UV bands at
262 nm and 295 nm of LamA were notably decreased
upon heat treatment at 110C, and significantly
sup-pressed after heat incubation at 150C or in the
pres-ence of 8 m GdnHCl The intensities of the bands
decrease when aromatic residues become more distant
from each other due to loose structure
The conclusions drawn from fluorescence
spectro-scopy are in line with CD The fluorescence emission
of native LamA showed maximum intensity at 335 nm
indicating moderate interaction of the tryptophans
with the solvent The emission profile of thermally
denatured LamA at 110C suggests that cooling to
room temperature did not result in refolding to the
native conformation but rather to a native-like form
The red-shift in the emission maximum indicates
increased tryptophan exposure It should be noted
that when the emission maxima are correlated with
tryptophan exposure to water, the interaction often
originates from penetration of water molecules into the
interior of the protein This is true especially for
struc-tural distortions induced by heat treatment In such
cases, the red-shift of the emission originates from
larger accessibility of tryptophans to both internal and external water The red-shift to 356 nm, which was observed upon heat incubation at 150C, suggests protein unfolding In the presence of 8 m GdnHCl the emission maximum was observed at 350 nm which is red-shifted as compared to the respective maximum
of the native state, but not as much as that of the unfolded protein
Analysis of the fluorescence properties of multitryp-tophan proteins is a difficult task even when the struc-ture is known The emission spectrum represents the average of local quenching and complicated resonance energy transfer phenomena Apart from the influence
of the polar solvent, which decreases the fluorescence emission of exposed tryptophans, in the protein matrix tryptophans can be quenched by neighboring carboxyl groups, histidine, methionine, phenylalanine, lysine, etc [22] Energy transfer from one tryptophan to another tryptophan or to a tyrosine decreases the fluorescence quantum yield of the donor [23] LamA has a single cysteine that is likely to play a critical role
as sulfhydryl groups are notorious quenchers of the proximal tryptophans [24] After thermal denaturation
at 110C, the fluorescence intensity moderately decreased while incubation at 150C resulted in sub-stantial three-fold decreased emission This observation and the red-shift of the emission maximum at 356 nm suggest that in this conformation, the tryptophans are quenched, possibly due to contact with water The residual intensity may imply that even in the case of
an extensively hydrated unstructured backbone it is possible that tryptophan(s) belong to a locally struc-tured domain The twofold increase of the fluorescence intensity in the presence of 8 m GdnHCl probably ori-ginates from relocation of tryptophans in the three dimensional structure of the protein In the new posi-tions the interacposi-tions of the tryptophans with quench-ing groups are weaker and⁄ or the intertryptophan distances are longer than that required for energy transfer [25,26] Both mechanisms increase the fluores-cence quantum yield, which overwhelms the quenching effect of the solvent-exposed tryptophans Hence, from both the emission maximum and the fluorescence intensity it is concluded that even at 8 m GdnHCl there is a residual structure in LamA that involves buried tryptophan residue(s)
Notable differences in LamA before and after ther-mal and chemical treatment were also observed upon interaction with ANS (Fig 4) The heat-denatured state is probably characterized by a structural distor-tion from which dissolved ANS accessed hydrophobic groups that were previously located in the interior of the protein [27] This interaction led to a significantly
Trang 9increased intensity The unfolded state upon
incuba-tion of LamA at 150C was justified by the blue-shift
of the ANS emission maximum
Time-resolved fluorescence gave insight into the
tryptophans’ relaxation dynamics Conformational
changes were justified by simple inspection of the
fluor-escence decays The fluorfluor-escence of the heat- and
chem-ically-treated LamA decayed at longer lifetimes This is
typical for proteins with solvent exposed tryptophans
[28] Each of the five (four in the presence of 8 m
GdnHCl), lifetimes, si, resolved represents a class of
tryptophans in a specific microenvironment [29–31],
and the respective pre-exponentials, ai, are related to
the fraction of tryptophans in each class [28,32,33] In
native LamA the extremely short lifetime (28 ps),
which accounts for one third of the total fluorescence
intensity, can be assigned to very efficient energy
trans-fer or to strong static quenching from amino acid(s)
(e.g cysteine) in the vicinity of the emitting
trypto-phans The amplitude of the longest lifetime, s5, in
native LamA at 5.5 ns probably represents
water-exposed tryptophans and contributes very little to the
total fluorescence The picture is reversed after heat
and chemical treatment, where the contribution from
the longest lifetimes is significantly increased
After heating at 110C the tryptophans,
character-ized by extremely short-lived relaxation in the native
state, were now decayed at a slightly longer lifetime
(37 ps) Notably, the amplitude, a1, of the tryptophans
emitting at the shortest lifetime is similar to that
resolved for native LamA The amplitude for
trypto-phans that decay at longer lifetimes was markedly
increased, which suggests that the slightly exposed
tryp-tophans of the native protein become more exposed in
the molten globule and therefore more
solvent-quenched
Studies in helical peptides and in small b-structured
proteins show that the amplitudes for each decay
com-ponent vary with the secondary structure [34,35] The
fluorescence from tryptophans belonging to an
exten-ded b-conformation decays with significant
contribu-tion from intermediate lifetimes This is the case for
native LamA Interestingly, the apparent increase of
a-helices and the decrease of sheets and strands upon
heat treatment at 110C and in the presence of 8 m
GdnHCl, as evidenced by far-UV CD, were confirmed
by the time-resolved fluorescence measurements: the
increased pre-exponential term a5 of the longest
decay time and the decreased contribution of the
intermediate components (i.e., a2–a4for the native and
a2–a3 for the heat and chemically treated LamA)
sug-gest decreased b-structures and increased helical
con-tent [34]
The contribution, ai, of the longest lifetime compo-nents (si> 3.8 ns) to the total fluorescence signifi-cantly increased from the native LamA to the thermally denatured at 110C LamA, to the heat unfolded LamA, to the chemically treated partially unfolded protein This order is consistent with the increased solvent exposure of the tryptophans in the heat-treated samples as observed in the steady-state fluorescence spectra There is a deviation from the order in the case of LamA in the presence of 8 m GdnHCl (Fig 5) This could be due to the significant contribution from completely exposed tryptophans (s5> 7 ns) of the heat unfolded protein that is absent
in the GdnHCl partially unfolded LamA However, we should bear in mind that steady-state measurements provide an intensity-weighted, time-averaged descrip-tion of the fluorophore emission and, hence, are pro-portional not to the most populated state but to the state that emits most This trait of steady-state emis-sion and the fact that specific interactions may elude time-resolved fluorescence detection and, thus, conceal
a part of the interpretation are additional reasons for the discrepancy
An analysis of the time-resolved anisotropy in terms
of protein conformer-lifetime assignments was also attempted The rapidly relaxing component in native LamA, /1, can be ascribed to flexibility of the indole ring in the protein matrix or other local dynamic events of the tryptophans which cause very fast de-polarization Upon heat and chemical treatment, the tryptophans rotate more freely as a result of rearrange-ments in the protein matrix around the fluorophore(s) This is shown in the fractional contribution b1 of the short correlation time and the calculated rotation angle
of the tryptophans in Table 2
The presence of many tryptophans distributed over the protein backbone is advantageous for the calcula-tion of the rotacalcula-tional diffusion of a protein in solucalcula-tion The rotational properties depend on the orientation of the dipoles relative to the main symmetry axis and, hence, a large number of fluorophores ensures that all orientations are sampled and the pristine rotational correlation time is determined by the anisotropy decay [36] After thermal denaturation at 110 C, the long lived component slightly increased to 19.39 ns In the completely and partially unfolded states the intra-molecular interactions and internal structural con-straints are loosened or lost and, hence, large parts
of the polypeptide chain become solvent exposed Therefore, the rotational freedom of the tryptophans substantially increases and the system loses anisotropy much faster (Fig 6; curves c and d) In these cases, the size of LamA could not be determined from the
Trang 10parameters recovered due to hydration of internal
pro-tein segments resulting in largely expanded
conforma-tions The medium correlation time of 3.8 ns that was
observed in the completely unfolded LamA
corres-ponds to tryptophans trapped locally that just lose
anisotropy faster than the tryptophans in the native
state (Table 3) The medium-lived component (2.1 ns)
observed in the guanidine-treated partially unfolded
LamA emerged at the expense of the shortest
pico-second correlation lifetime Motions with correlation
times ranging from 1 to 3 ns describe segmental
back-bone fluctuations of the polypeptide chain [37,38]
These motions are important when the protein
integ-rity is disrupted and the protein backbone is solvated
and more flexible
Data from CD, fluorescence spectroscopy
(steady-state, time-resolved and ANS binding), and anisotropy
were used to probe conformational features of LamA
before and after heat or chemical treatment It was
suggested that upon heating at 110C, the local
micro-environment of the tryptophans resembles but it is not
identical to that of the native state It is likely that this
state represents a structurally disturbed or locally
unfolded state rather than completely unfolded The
structural elements may be maintained by a
mechan-ism involving specific local and long-range interactions,
some of which are native-like [39–43] The interaction
of LamA with 8 m GdnHCl resulted in significant
structural changes but not in complete unfolding The
protein was partially unfolded with characteristics
clearly distinct from the completely unfolded
confor-mation obtained after incubation at 150C
Experimental procedures
Purification of LamA, treatment and chemicals
The gene encoding LamA (sequence deposited in GenBank:
accession No AF013169) was isolated from P furiosus and
after cloning it was overexpressed in Escherichia coli BL21
(DE3) using the T7 expression system [3] Further
purifica-tion was achieved by size exclusion chromatography in a
Superdex 200 column (Amersham Pharmacia, Piscataway,
sodium phosphate buffer at pH 7.0 The protein
concentra-tion was routinely determined by the absorpconcentra-tion at 280 nm
Controlled heat treatment of LamA was carried out in a
VP-DSC calorimeter (MicroCal Inc., Northampton, MA,
tem-perature and used for further analyses Heat incubation for
con-trolled oil bath using thick-walled glass tubes with a lid
capable of withstanding the vapor pressure of water Chem-ical denaturation was studied in the presence of extra pure fluorescence-free GdnHCl (Merck, Rahway, NJ, USA) The GdnHCl solutions were prepared according to Pace et al [44] and the concentration was determined by measuring their refractive index LamA was allowed to interact with
Circular dichroism
1 mm and 1 cm quartz cuvettes, respectively, were recorded
in a JASCO J-715 (Tokyo, Japan) spectrophotometer equipped with a temperature controller (JASCO PTC 348
metal-caged quartz cuvette under pressure to prevent eva-poration of water The CD spectra referring to LamA
sam-ples which had been previously heated and then cooled to room temperature The spectrophotometer was calibrated
resolution, and 0.25 s response time Spectra of LamA before and after heat or chemical treatment resulted from accumulation of 32 scans that were subsequently aver-aged Blank spectra of buffer without protein, obtained at identical conditions, were subtracted Data analysis was performed by fitting the acquired spectra with reference spectra using the contin program, which is based on nonlinear regression fitting algorithms without constraints (ridge-regression analysis) [45,46] This program gives a much better estimate of b-sheets and turns than simple multiple linear regression [47] An average molar mass of
115 Da per amino acid residue was used for calculating the ellipticity, h
Steady-state fluorescence spectroscopy Fluorescence emission was measured by a Varian Cary Eclipse spectrophotometer (Variam, Palo Alto, CA, USA) Unless otherwise indicated, all measurements were carried
the range 300–400 nm on excitation at 300 nm The excita-tion and emission slit widths were set at 5.0 and 2.5 nm, respectively All spectra were corrected for the background emission of water Spectra of samples containing GdnHCl were corrected using as reference the buffer solution with the same concentration of GdnHCl Binding of ANS was studied between 400 and 600 nm on excitation at 380 nm
heat and chemical treatment