Characterizationofmonomericsubstatesof ascorbate
oxidase
Almerinda Di Venere
1,2,
*, Eleonora Nicolai
1,2,
*, Nicola Rosato
1,2
, Antonello Rossi
1,2
, Alessandro
Finazzi Agro
`
2
and Giampiero Mei
1,2
1 NAST Centre, University of Rome, ‘Tor Vergata’, Italy
2 Department of Experimental Medicine and Biochemical Sciences, University of Rome, ‘Tor Vergata’, Italy
Introduction
The folding of oligomeric enzymes is a fascinating, but
still poorly understood, process. Obviously, the
sequence of events that drive the polypeptidic chains
to their final quaternary structure may differ from pro-
tein to protein. However, all nascent oligomers should,
sooner or later, face the same crucial step, i.e. the for-
mation of a suitable interface al lowi ng the mutual recog-
nition and the sticking together of the subunits [1,2].
This step is often performed after the partially folded
monomers are assembled. The identification and char-
acterization of these intermediates are unfortunately
quite difficult, even in vitro, as the new synthesized
monomers can be unstable, even under mild denatur-
ant conditions. As a matter of fact, several unfolding
studies on dimeric proteins published so far show a
simple two-state equilibrium model, i.e. a direct transi-
tion from unfolded monomers to folded dimers [3–5].
Thus, the lack of detectable monomeric intermediates
does not allow discrimination between folding and
oligomerization. Nonetheless, when stable monomers
have been identified along the folding pathway, many
more details of their specific structural properties are
becoming apparent. In a few cases, both monomeric
and dimeric intermediates have been found to be pres-
ent in the folding landscape of dimeric proteins [6,7].
For example, in the case ofascorbateoxidase (ascorbic
Keywords
dimeric proteins; fluorescence correlation
spectroscopy; folding intermediates;
high pressure
Correspondence
G. Mei, Department of Experimental
Medicine and Biochemical Sciences,
University of Rome, ‘Tor Vergata’,
Via Montpellier 1, 00133 Rome, Italy
Fax: +39 06 72596468
Tel: +39 06 72596460
E-mail: mei@med.uniroma2.it
*These authors contributed equally to this
work
(Received 30 November 2010, revised 10
February 2011, accepted 23 February 2011)
doi:10.1111/j.1742-4658.2011.08084.x
Ascorbate oxidase (AAO) is a large, multidomain, dimeric protein whose
folding ⁄ unfolding pathway is characterized by a complex, multistep pro-
cess. Here we used fluorescence correlation spectroscopy to demonstrate
the formation of partially folded monomers by pH-induced full dissociation
into subunits. Hence, the structural features ofmonomeric AAO could be
studied by fluorescence and CD spectroscopy. We found that the mono-
mers keep their secondary structure, whereas subtle conformational
changes in the tertiary structure become apparent. AAO dissociation has
also been studied when unfolding the protein by high hydrostatic pressure
at different pH values. A strong protein concentration dependence was
observed at pH 8, whereas the enzyme was either monomeric or dimeric at
pH 10 and 6, respectively. The calculated volume change associated with
the unfolding ofmonomeric AAO, DV )55 mLÆmol
)1
, is in the range
observed for most proteins of the same size. These findings demonstrate
that partially folded monomeric species might populate the energy land-
scape of AAO and that the overall AAO stability is crucially controlled by
a few quaternary interactions at the subunits’ interface.
Abbreviations
AAO, ascorbate oxidase; ANS, 1-anilino-8-naphthalenesulfonic acid; CD, circular dichroism; FCS, fluorescence correlation spectroscopy.
FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS 1585
acid oxidase, AAO, EC 1.10.3.3), a combination of
hydrostatic pressure and low denaturant concentrations
led to the following scheme for the unfolding process:
N
2
$ MD $ 2M
$ 2U
where the native dimeric enzyme (N
2
) obtains the
fully unfolded monomeric form (U) through a series of
partially unfolded dimeric (MD) and monomeric (M*)
states [8].
At pH 6–7, AAO is a large, stable homodimer (rela-
tive molecular mass = 140 kDa). It catalyses the
reaction between oxygen and ascorbate, yielding
dehydroascorbate and water [9]. Each subunit is
formed by three distinct domains and contains four
copper ions, three of which are located at the interface
between domains, forming a so-called trinuclear centre
[10,11]. Previous equilibrium unfolding studies [8,12]
have suggested that salt bridges and electrostatic inter-
actions occurring at the dimeric interface play a crucial
role in the stabilization of the monomer’s tertiary
structure. These findings lead us to conjecture that the
monomer–monomer interaction could be weakened at
certain pH values.
Therefore, we used fluorescence correlation spectros-
copy (FCS) to demonstrate that the protein can be
fully dissociated into monomers at alkaline pH. The
dependence of the dissociation process on pH and pro-
tein concentration was studied by applying an external
physical stress to the protein structure, i.e. by raising
the hydrostatic pressure. Furthermore, the stability of
the monomers was also studied as a function of urea
and guanidine concentration and their structural fea-
tures characterized by circular dichroism (CD), absorp-
tion and fluorescence spectroscopy.
Results
AAO is fully dissociated into monomers at
pH = 10
Early studies on AAO have given evidence that AAO
oligomerization is strongly pH dependent, the dissocia-
tion into monomers being progressively promoted in
the alkaline range [13]. Here we used several spectro-
scopic techniques to monitor the quaternary structure
of AAO at different pH values. In the first set of
experiments, the protein was dansylated and its anisot-
ropy measured, in order to obtain information on the
AAO rotational motion in solution. Dansyl chloride is
particularly suited to obtaining information on the
rotational diffusion of large proteins in solution, due
to its long fluorescence lifetime (10 ns). When cova-
lently bound to the terminal NH
2
of AAO, the dansyl
fluorescence anisotropy was r 0.152. This value
progressively decreased to 0.130 upon alkalinization of
the sample, indicating the occurrence of a faster rota-
tion of the molecule (Fig. 1). An almost full recovery
of the initial anisotropy (> 94%) was obtained by
bringing the sample back to pH 7.0 (Fig. 1A).
Assuming a spherical shape, the estimated rotational
correlation times, / ¼
gV
RT
, from a Perrin plot (Fig. 1A,
0.130
0.135
0.140
0.145
0.150
0.155
6789101112
<r >
pH
9
11
13
15
17
10
000 100 000
Ve (mL)
Molecular weight (Da)
-lactalbumin
Carbonic anhydrase
Chicken egg albumin
BSA
AAO pH10
AAO pH6
Urease
A
B
Fig. 1. (A) Steady-state anisotropy of dansylated AAO at increasing
(filled symbols) or decreasing (empty symbols) pH values. Excita-
tion was k
exc
= 336 nm, whereas emission was collected through
a LG420 cut-off filter. Inset: Perrin plot reported at pH 6 (empty
squares) and pH 10 (filled circles). (B) AAO relative molecular mass
determination by gel filtration chromatography at two different pH
values.
Monomeric intermediates of AAO A. Di Venere et al.
1586 FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS
inset), were 37 and 60 ns for the monomer (pH = 10)
and the dimer (pH = 6), respectively [14].
The corresponding hydrated volume of the protein
molecule at pH 10 was 1.5 · 10
5
A
˚
3
, reasonably
close to the size of the AAO monomer in the crystal,
that is = 49 · 53 · 65 A
˚
3
[9]. Size exclusion chromato-
graphy at pH 10 (Fig. 1B) yielded an elution volume
close to that of bovine serum albumin (relative molec-
ular mass 68 kDa), thus confirming the presence of
monomers in alkaline buffer.
These results provide evidence that at high pH val-
ues the subunits of AAO can be dissociated. However,
a direct evaluation of the dissociation extent in an
alkaline environment was still required. In order to
obtain this information, we performed FCS measure-
ments on fluorescein-labelled AAO at different pH
values, in the range 7.2–10. Measurements of fluo-
rescein alone in buffer yielded a diffusion coefficient
D
fluorescein
290 lm
2
Æs
)1
, in good agreement with the
value reported in the literature [15]. The diffusion coef-
ficient for fluorescein-labelled AAO measured at pH
7.2 was D
AAO
46 lm
2
Æs
)1
, corresponding to a rela-
tive molecular mass of 149 kDa [16], matching the
size of dimeric AAO [10,11].
The FCS autocorrelation curves of AAO are shown
in Fig. 2, together with the best fits corresponding to
the minimum v
2
value. Increasing the pH induces a
progressive decrease in the fluctuation amplitude G(0),
which represents the value of the autocorrelation func-
tion at s = 0. The inverse of this parameter, evaluated
by fitting the experimental data, is proportional to the
overall particle concentration [17] from which the frac-
tion of monomers can be extrapolated. The results
(Fig. 2, inset) demonstrate that the number of fluores-
cent AAO molecules doubled at pH = 10, thus con-
firming that a full dissociation into monomers was
achieved.
Characterization of AAO monomer structural
features
The secondary and tertiary structures of monomeric
AAO were investigated by CD, absorption and fluo-
rescence spectroscopy. As shown in Fig. 3A, a loss of
the CD signal at 260, 288 and 295 nm was
observed, indicating a reduced asymmetry in the envi-
ronment of the aromatic side chains [18]. Instead the
CD spectra at pH 6 and pH 10 are quite superimpos-
able in the peptidic region (Fig. 3A, inset), indicating
that AAO retained its overall secondary structure
upon subunit dissociation. The broadening of the fluo-
rescence emission spectrum (Fig. 3B) and the decrease
in the absorption band at 610 nm (Fig. 3B, inset) con-
firm that significant changes were taking place at the
level of the protein tertiary structure, affecting the
microenvironment of tryptophans [19] and copper.
This result is indeed compatible with the faster local
dynamics suggested by the fluorescence decay measure-
ments (Fig. 3C). In particular, the broadening of the
longer lifetime component and its shorter mean life-
time value indicate that both structural heterogeneity
and quenching effects were greater at pH 10, as
expected for an overall less compact tridimensional
structure. This feature has been confirmed by 1-anili-
no-8-naphthalenesulfonic acid (ANS) binding. Indeed,
an increase in the fluorescence intensity and a blue
shift in the emission of ANS was observed in the pres-
ence of AAO at pH 10, indicating that the monomers
are characterized by a greater exposure of ANS-bind-
ing protein surface with respect to the dimeric
structure at pH 6.
Stability of AAO monomer
The stability of AAO monomers at pH 10 was studied
in the presence of denaturants or under pressure. In
the first case, urea or guanidinium hydrochloride was
used to induce protein unfolding, which was monitored
by CD (signal intensity at 220 nm) and steady-state
fluorescence measurements (fluorescence intensity at
350 nm).
0.00
0.02
0.04
0.06
0.08
10
–5
10
–4
10
–3
10
–2
10
–1
G( )
(s)
0
10
20
30
78910
pH
[AAO] n
M
Fig. 2. Autocorrelation curves of fluorescein-labelled AAO at differ-
ent pH values, namely 7.2 (red), 8.7 (orange), 9.0 (green), 9.6 (blue),
10.0 (purple). Inset: extrapolated overall concentration of AAO
molecules at each pH value.
A. Di Venere et al. Monomeric intermediates of AAO
FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS 1587
The normalized data (shown in Fig. 4A) demon-
strated that both denaturants induced a sigmoidal
transition, not dependent on protein concentration,
indicating the absence of a monomer–dimer equilib-
rium at pH 10. Furthermore, the two spectroscopic
techniques gave superimposable curves (Fig. 4A, inset),
suggesting the occurrence of a simple two-state transi-
tion, from the native to the unfolded state. The data
were therefore fitted according to this model and the
results are reported in Table 1 (rows 5–6). When com-
pared with the overall free energy change
(DG > 16 kcalÆmol
)1
) previously obtained [12] for the
denaturation of the dimeric enzyme (Table 1, rows
1–4), the free energy of unfolding (DG 3 kcalÆmol
)1
)
indicates that the AAO subunits were only marginally
stable.
The effect of hydrostatic pressure on the AAO
monomeric species was monitored by steady-state fluo-
rescence measurements, recording the emission spec-
trum in the range 1–3000 bar. In Fig. 4B the red
shift of the spectral centre of mass due to the progres-
sive exposure of the tryptophan residues to the solvent
is shown. The data clearly demonstrate that at higher
pH values a larger hydration was achieved. Further-
more, AAO concentration strongly affected the transi-
tion recorded at pH 8, but not at pH 6 or pH 10,
where the protein was in the dimeric and monomeric
form, respectively (Fig. 1B, 2). In order to compare
the strength of mechanical unfolding with chemical
denaturation of the monomeric species, the data at pH
10 were fitted according to a simple two-state transi-
tion and the parameters corresponding to the best
two-state fit reported in Table 1 (row 7). Similar
results were obtained using fluorescence intensity (data
not shown).
Discussion
The detection of a monomer–dimer equilibrium occur-
ring during the folding process of large oligomeric
enzymes is expected to be complex due to the simulta-
neous presence of molecular species with different sizes
0.00
0.10
0.20
0.30
0.40
250 260 270 280 290 300
[ ] x 10
–3
(deg cm
2
·dmol
–1
)
Wavelength (nm)
0
20
40
60
80
100
300 340 380 420
Fluorescence (a.u.)
Wavelength (nm)
x 2
0
0.2
0.4
0.6
024
Fraction
Lifetime (ns)
A
B
C
Fig. 3. Spectroscopic features of AAO at pH 6 (circles) and pH 10
(solid line). (A) CD spectra in the aromatic and peptide (inset)
regions. (B) Steady-state fluorescence spectra of AAO tryptophan
residues (solid line = pH 10; circles pH = 6). The dashed line repre-
sents the difference spectrum. Inset: respective absorption spectra
in the visible spectrum. (C) Lifetime distribution profiles obtained
from dynamic fluorescence measurements (k
exc
= 300 nm; emis-
sion through a WG 320 cut-off filter) at pH 10 and 6 (solid line and
circles, respectively). Inset: ANS spectra in the presence of AAO at
the same pH values.
Monomeric intermediates of AAO A. Di Venere et al.
1588 FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS
and shapes [20]. In particular, an accurate evaluation
of the extent of the subunit aggregation process is
often impossible using traditional biophysical tech-
niques, as they are generally based on the detection of
an overall macroscopic signal, which is a nontrivial
combination of the spectroscopic features of each indi-
vidual species. In the last 10 years, several studies on
oligomers have demonstrated that FCS can easily cir-
cumvent this problem, by simply ‘counting’ the num-
ber of particles present in the volume explored. For
instance, FCS has been used to follow protein trans-
portation in axons [21], to detect enzyme aggregation
states in the presence of lipid vesicles [22] or to moni-
tor the oligomerization of proteins involved in receptor
binding [23] or in neurodegenerative diseases [24].
The characterizationof the individual properties of
monomeric and dimeric species is crucial to unravel
which kinds of interaction drive an oligomer folding
process. Using the combination of high pressure and
chemical denaturants, we have already demonstrated
the presence of both monomers and dimers in the
unfolding transition of AAO [8,12]. However, although
the structural–functional features of the dimeric inter-
mediate have been characterized in detail [12], no
information is available as yet on the structure of the
subunits. In fact, the low stability of monomers
( 2 kcalÆmol
)1
) and the experimental set-up used (i.e.
a pressurized cell, [11]) prevented their isolation out of
the various species populating the folding pathway
of AAO. Here we have shown that stable monomeric
0.0
0.2
0.4
0.6
0.8
1.0
02468
[denaturant] (
M)
Fraction of unfolded
334
336
338
340
0 1000 2000 3000
Center of mass (nm)
Pressure (bar)
A
B
Fig. 4. (A) Fluorescence unfolding profiles ofmonomeric AAO (at
pH 10, 2 l
M), using guanidinium hydrochloride (cyan circles) or urea
(red squares). Black symbols correspond to the same experiments
at 20 times lower AAO concentration. The solid lines represent the
best fits of the more concentrated AAO unfolding transition. Inset:
superposition of the fraction of guanidinium hydrochloride-unfolded
species obtained from CD (green squares) and fluorescence (cyan
circles) data. Black squares represent the fraction of refolded AAO
molecules obtained by CD measurements. (B) Red shift of the
AAO (monomer concentration 2 l
M) emission spectrum (reported
in terms of the spectral centre of mass) upon pressure at three dif-
ferent pH values, namely pH 6 (green circles), pH 8 (red triangles)
and pH 10 (cyan circles). The black symbols correspond to the
same experiments at 20 times lower AAO concentration. Inset: CD
spectra at pH 10, 1 bar, before (cyan) and after (black) a pressuriza-
tion cycle.
Table 1. Thermodynamic parameters of the AAO unfolding transi-
tions
a
.
Sample
DG
[kcalÆmol
)1
]
m
[kcalÆmol
)1
ÆM]
DV
[mLÆmol
)1
]
N
2
() MD (P,G,U)
b
Fig. 5, transition 1
3.1 ± 0.4 )63 ± 4
MD () 2U (G,U)
b
Fig. 5, transition 2
13.0 ± 0.7 1.1 ± 0.2 (U)
1.7 ± 0.2 (G)
MD () 2M (P)
b
Fig. 5, transition 3
10.7 ± 0.5 )171 ± 15
M () U (P)
b
Fig. 5, transition 4
2.0 ± 0.2 )54 ± 5
M () U (U)
[AAO] 2 l
M
[AAO] 0.1 lM
Fig. 5, transition 4
3.0 ± 0.2
2.9 ± 0.2
0.8 ± 0.1
0.7 ± 0.1
M () U (G)
[AAO] 2 l
M
[AAO] 0.1 lM
Fig. 5, transition 4
3.3 ± 0.2
3.1 ± 0.3
1.7 ± 0.1
1.7 ± 0.1
M () U (P)
Fig. 5, transition 4
2.2 ± 0.2 )55 ± 8
a
Parameters have been obtained by fitting unfolding transition
curves as a function of pressure (P), guanidinium hydrochloride (G)
and urea (U).
b
Average values from [8,12]. N
2
, native dimeric
enzyme; U, unfolded monomeric form; MD, partially unfolded
dimeric state; M, monomeric state.
A. Di Venere et al. Monomeric intermediates of AAO
FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS 1589
species (Fig. 1A, B) might be obtained at pH 10 and
the full dissociation into monomers can be checked by
counting the number of particles through the FCS
technique (Fig. 2). The analysis of absorption, CD and
fluorescence spectra at pH 10 demonstrate that the
monomerization induces conformational changes that
involve the protein tertiary structure, without affecting
its secondary structure (Fig. 3A, B). This feature could
recall the properties of the molten globule state [25],
which is often present in the folding pathway of globu-
lar proteins. Both dynamic fluorescence and ANS
binding measurements support such a hypothesis, indi-
cating an enhanced heterogeneity experienced by the
tryptophan residues (Fig. 3C) and a partial exposure
of hydrophobic patches on the protein surface
(Fig. 3C, inset), in the monomeric state at pH 10. The
abrupt decrease in enzymatic activity observed above
pH 8 [26] might be another consequence of subunit
dissociation, as suggested by the significant decrease
( )30%) in the AAO characteristic absorption band
at 610 nm (Fig. 3B, inset). This band is associated with
the so-called ‘blue copper’ (or type I copper), located
in domain 3 of AAO, which takes up electrons from
ascorbate [10].
Despite these significant modifications of the ter-
tiary structure, upon pressurization the dissociated
subunits showed a volume change consistent with
that found for the full denaturation of most mono-
meric proteins [27], suggesting that AAO monomers
were retaining most of the native interactions. It
is worth mentioning that a similar DV ( )50 ⁄ )60
mLÆmol
)1
) for the unfolding of each AAO subunit
was already indirectly obtained studying the stability
of the partially folded dimeric intermediate [8] that
populates the protein folding pathway. All these
findings allow a graphic representation to be deter-
mined of the different folding ⁄ unfolding pathways of
AAO explored so far (Fig. 5). Interestingly, the low
stabilization energy of monomers and their signifi-
cantly heterogeneous conformational substates (as
probed by the tryptophan residues, Fig. 3C) make
them more related to the unfolded state than to the
monomers assembled in the dimeric AAO, again
stressing the role of dimerization on the stability of
many proteins [3]. However, the monomers still
retain most of their native tridimensional structure,
as indicated by the spectral properties reported
above (Figs 3A, B). Such a paradox might be recon-
ciled by assuming a weakening of electrostatic inter-
actions among the three domains of each AAO
subunit, once the quaternary structure is lost.
Indeed, the X-ray crystallographic model of AAO
(Fig. 6) has shown the presence of a huge number
of intrasubunit ion pairs (42), one-third of which are
located at the interdomain surfaces [11]. Previous
equilibrium unfolding measurements [12] have dem-
onstrated that 16.1 kcalÆmol
)1
are needed to stabilize
the native AAO molecule at pH 6.8, whereas pres-
sure-induced dissociation of the dimeric intermediate
yielded 10.7 kcalÆmol
)1
(Table 1). Thus, it can be
argued that more than 60% (10.7⁄ 16.1) of the over-
all stabilization energy can be attributed to quater-
nary interactions. Actually, such an evaluation is an
overestimate, as the presence of fully unfolded
monomers ( 15%) was detected even at a low
denaturant concentration [12]. However, a lower
limit can also be obtained taking into account the
free energy of unfolding of the monomers at pH 10
( 3 kcalÆmol
)1
, Table 1). In particular, considering
that the DG° values are referred to 1 mole of parti-
cles, the fractional contribution of the monomers
must be around 37% (i.e. 3 · 2 ⁄ 16.1). In this case,
subtracting the 3.1 kcalÆmol
)1
of the first transition
(N
2
() MD), the contribution of the quater-
nary structure to the overall stability is 43%
(16.1 ) 3.1 ) 3 · 2) ⁄ 16.1. In conclusion, these
N
2
MD
U
M
1
2
3
4
5
Fig. 5. Schematic three-dimensional representation of the possible
folding ⁄ unfolding pathways of AAO. The depth and width of each
potential well are respectively proportional to the free energy
change required to reach the fully unfolded state, and to the extent
of tertiary structure heterogeneity (as estimated from the widths of
the lifetime distributions). The arrows correspond to the transition
induced by denaturant(s), pressure or pH, as specified in Table 1.
N
2
, native dimeric enzyme; U, unfolded monomeric form; MD,
partially unfolded dimeric state; M, monomeric state.
Monomeric intermediates of AAO A. Di Venere et al.
1590 FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS
estimations provide an upper (60%) and a lower
(44%) limit to the contribution of the quaternary
structure to AAO stability. Considering the small
size of the AAO dimeric interface ( 1100 A
˚
2
) and
the large gap separating the two subunits [8], this is
certainly an important effect. It follows that a few
contacts at the dimeric interface are sufficient to
control the overall stability, preventing solvent acces-
sibility to the interdomain surfaces. Such a feature
is not unique to the AAO quaternary structure,
but appears to be a common folding strategy
for large dimers [3], which generally lack stable
and folded monomeric intermediates in their folding
pathway.
It would be worthwhile to extend similar studies to
other dimeric proteins in order to understand whether
dimerization is a universal way of stabilizing mono-
meric proteins that would otherwise be unstable. Of
particular interest would also be to determine if the
dimerization can occur very early, during the polypep-
tide chain synthesis at the polyribosomes.
Materials and methods
Materials
Ultrapure urea, guanidinium hydrochloride, glycerol and
dansyl chloride were purchased from Sigma (St Louis,
MO, USA); AAO (EC 1.10.3.3) from green zucchini from
Boehringer Mannheim (Mannheim, Germany); ANS and
fluorescein from Molecular Probes (Eugene, OR, USA).
The enzyme was dissolved in different buffers depending
on the pH value required. Between pH 6 and pH 8 the
buffer used was potassium phosphate, except for pressure
experiments, in which Bistris ⁄ HCl 20 mm (pH = 6.0) and
Tris ⁄ HCl 20 mm (pH = 8.0) were used. For pH values
‡ 9.0, AAO was always dissolved in a 0.2 m
NaCO
3
⁄ 10 mm EDTA buffer.
Dansylation of AAO
Dansyl chloride, dissolved in acetone, was added to the
protein solution at pH = 8.0 in a ratio 50 : 1. After a 24 h
incubation (in the dark, at 4 °C) the excess dansyl was
removed by size exclusion chromatography in a D-salt gel
(cut-off = 5000 Da), equilibrated at pH = 8.0. The final
ratio dansyl chloride ⁄ AAO was determined spectrophoto-
metrically measuring the absorbance of the sample at 280
and 340 nm, to estimate the protein (280 nm) and the dan-
syl chloride (340 nm) concentration. All our labelled sam-
ples had a ratio of 0.7 and were fully active with respect
to native enzyme at pH 8.0.
Size exclusion column chromatography analysis
of AAO
A 31 cm column with a diameter of 9 mm was packed with
Sephacryl S200 HR (Amersham Pharmacia Biotech, Piscata-
way, NJ, USA) according to the manufacturer’s instructions.
The column was equilibrated and run with 80 mm potas-
sium phosphate buffer at pH 6.0 or 0.2 m NaCO
3
⁄ 10 mm
EDTA buffer at pH 10. The typical flow rate was 0.2
mLÆmin
)1
.
Calibration was carried out by running reconstituted
alpha-lactalbumin (bovine milk, molecular mass =
14 200 Da), carbonic anhydrase (bovine erythrocites,
molecular mass = 29 000 Da), albumin (chicken egg, molec-
ular mass = 45 000 Da), albumin (bovine serum, molecular
mass = 66 000 Da) and urease (jack bean, molecular
mass = 272 000 Da) (all from Sigma).
For AAO analysis, 0.5 mL purified samples, either at pH
6.0 or 10.0, were loaded.
Fluorescence, absorption and CD spectroscopy
Steady-state fluorescence spectra and anisotropy were
recorded on a K2-ISS photon counting fluorometer (ISS,
Champaign, IL, USA) equipped with Glan Thompson
polarizers. Excitation was set at 292 or 360 nm for intrinsic
and dansyl-labelled AAO fluorescence, respectively. Perrin
plots were obtained varying the temperature or the viscosity
of the solution with glycerol.
Dynamic fluorescence measurements of AAO 2 lm ,at
pH 6 or pH 10, were performed with a KOALA-ISS fluo-
rometer, using the phase shift and demodulation technique.
The excitation source (300 nm) was a laser diode; emission
was collected through a 320 WG cut-off filter to avoid scat-
tered light. The data were fitted according to a double
lorentzian-shaped continuous distribution of lifetime in
Fig. 6. Three-dimensional structure of AAO (Protein Data Bank
code 1aoz). The three domains of each subunit are shown in red,
blue and green; the amino acids at the subunit interface are repre-
sented in magenta.
A. Di Venere et al. Monomeric intermediates of AAO
FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS 1591
view of the heterogeneous and complex decay of the AAO
molecule, as described elsewhere [28].
Absorption was measured with a Perkin-Elmer Lambda-
18 spectrophotometer. CD spectra were collected with a
Jasco J-700 spectropolarimeter using a 0.1 and 0.5 cm path
length quartz cell, in the peptidic and aromatic regions,
respectively.
FCS measurements
Experiments were performed with the ISS-ALBA fluores-
cence correlation spectrometer equipped with a Nikon
inverted microscope. Two-photon excitation (in the range
780–800 nm) was provided by a Ti:sapphire mode-locked
laser (Chameleon Ultra; Coherent Inc, Santa Clara, CA,
USA). The instrument alignment was performed using a
dilute solution ( 10 nm) of 6G rhodamine. At each pH
value, a preliminary measurement with rhodamine, at a
known concentration, was carried out in order to
evaluate the excitation volume to be used in the data
analysis. Absorption measurements at 280 and 490 nm
were used to estimate the extent of AAO labelling.
A typical ratio fluorescein ⁄ AAO 0.6 was used in FCS
measurements. The data were analysed using iss-vista
software, fitting the points of the autocorrelation function
assuming a Gaussian–Lorentzian intensity profile distribu-
tion [29].
Equilibrium unfolding and pressure
measurements
Equilibrium unfolding measurements of the monomers were
performed after 24 h incubation at 4 °C in the presence of
different amounts of denaturants, or upon increasing the
hydrostatic pressure in the range 1–3000 bar, using the
high-pressure ISS cell device.
Refolding from the fully unfolded state was achieved
either by dialysis or by diluting the denaturants to the
desired final concentration.
The analysis of the unfolding measurements was per-
formed assuming a two-state transition model, MMU,
between the native (M) and the unfolded (U) state. The
equilibrium constant of the unfolding process, K, was sup-
posed to exponentially depend on the denaturant concen-
tration [30] according to:
K ¼ e
DG
0
þ m den½
RT
High-pressure measurements were performed with the
ISS K2 spectrofluorometer equipped with the ISS high-
pressure cell. The centre of mass of the fluorescence
steady-state emission spectra of AAO at pH 10 was fitted
according to a two-state model and an estimation of the
free energy of unfolding and partial molar volume change
obtained using the relationship [31]:
K ¼ e
DG
0
þ DVP½
RT
:
The error bars each data point in Figs 1A, 4A, B and
with each histogram of Fig. 2 (inset) were obtained by eval-
uating the standard deviation of three to four independent
measurements.
Acknowledgements
We thank Dr Beniamino Barbieri and Dr Shih-Chu
Liao for technical assistance with ISS instrumentation.
This study was supported in part by a grant from Min-
istero dell’Istruzione, dell’Universita
`
e della Ricerca
(PRIN 2008).
References
1 Rumfeldt JA, Galvagnion C, Vassall KA & Meiering
EM (2008) Conformational stability and folding mecha-
nisms of dimeric proteins. Prog Biophys Mol Biol 98,
61–84.
2 Neet KE & Timm DE (1994) Conformational stability
of dimeric proteins: quantitative studies by equilibrium
denaturation. Protein Sci 3, 2167–2174.
3 Mei G, Di Venere A, Rosato N & Finazzi-Agro
`
A
(2005) The importance of being dimeric. FEBS J 272,
16–27.
4 Marianayagam NJ, Sunde M & Matthews JM (2004)
The power of two: protein dimerization in biology.
Trends Biochem Sci 29, 618–625.
5 White GWN, Gianni S, Grossmann JG, Jemth P,
Fersht AR & Daggett V (2005) Simulation and experi-
ment conspire to reveal cryptic intermediates and a slide
from the nucleation-condensation to framework mecha-
nism of folding. J Mol Biol 350, 757–775.
6 Bose K & Clark AC (2001) Dimeric procaspase-3
unfolds via a four-state equilibrium process. Biochemis-
try 40, 14236–14242.
7 Cha
´
nez-Ca
´
rdenas ME, Pe
´
rez-Herna
´
ndez G, Sa
´
nchez-
Rebollar BG, Costas M & Va
´
zquez-Contreras E (2005)
Reversible equilibrium unfolding of triosephosphate
isomerase from Trypanosoma cruzi in guanidinium
hydrochloride involves stable dimeric and monomeric
intermediates. Biochemistry 44, 10883–10892.
8 Nicolai E, Di Venere A, Rosato N, Rossi A, Finazzi
Agro’ A & Mei G (2006) Physico-chemical properties of
ascorbate oxidase. FEBS J 273, 5194–5204.
9 Avigliano L & Finazzi-Agro’ A (1997) Biological Func-
tion and Enzyme Kinetics ofAscorbate Oxidase. In
Multi-copper Oxidase (Messerschmidt A ed), pp 265–
278. World Scientific, Singapore.
10 Messerschmidt A, Rossi A, Ladenstein R, Huber R,
Bolognesi M, Gatti G, Marchesini A, Petruzzelli R &
Finazzi-Agro
´
A (1989) X-ray crystal structure of the
Monomeric intermediates of AAO A. Di Venere et al.
1592 FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS
blue oxidaseascorbateoxidase from zucchini. Analysis
of the polypeptide fold and a model of the copper sites
and ligands. J Mol Biol 206, 513–529.
11 Messerschmidt A, Steigemann W, Huber R, Lang G &
Kroneck PM (1992) X-ray crystallographic characterisa-
tion of type-2-depleted ascorbateoxidase from zucchini.
Eur J Biochem 209, 597–602.
12 Mei G, Di Venere A, Buganza M, Vecchini P,
Rosato N & Finazzi-Agro’ A (1997) Role of quaternary
structure in the stability of dimeric proteins: the case
of ascorbate oxidase. Biochemistry 36, 10917–10922.
13 Avigliano L, Vecchini P, Sirianni P, Marcozzi G,
Marchesini A & Mondovi B (1983) A reinvestigation
on the quaternary structure ofascorbateoxidase from
Cucurbita pepo medullosa. Mol Cell Biochem 56,
107–112.
14 Cantor CR & Schimmel PS (1980) Biophysical Chemis-
try, Vol. II. Freeman, San Francisco, CA, pp. 459–465.
15 Sa
´
nchez SA, Brunet JE, Jameson DM, Lagos R &
Monasterio O (2004) Tubulin equilibrium unfolding fol-
lowed by time-resolved fluorescence and fluorescence
correlation spectroscopy. Protein Sci 13, 81–88.
16 Young ME, Carroad PA & Bell RL (1980) Estimation
of diffusion coefficient of proteins. Biotechnol Bioeng
22, 947–955.
17 Thompson NL (1991) Fluorescence correlation spectros-
copy. In Topics in Fluorescence Spectroscopy (Lakowicz
JR ed), pp 337–378. Plenum Press, New York.
18 Woody RW & Dunker AK (1996) Aromatic and cystine
side-chain CD in proteins. In Circular Dichroism and
the Conformational Analysis of Biomolecules (Fasman
GD ed), pp 109–157. Plenum Press, New York.
19 Reshetnyak YK, Koshevnik Y & Burstein EA (2001)
Decomposition of protein tryptophan fluorescence spec-
tra into log-normal components. III. Correlation
between fluorescence and microenvironment parameters
of individual tryptophan residues. Biophys J 81, 1735–
1758.
20 Galvagnion C, Smith MT, Broom A, Vassall KA, Meg-
lei G, Gaspar JA, Stathopulos PB, Cheyne B & Meier-
ing EM (2009) Folding and association of thermophilic
dimeric and trimeric DsrEFH proteins: Tm0979 and
Mth1 491. Biochemistry 48, 2891–2906.
21 Terada S, Masataka K & Hirokawa N (2000) Oligo-
meric tubulin in large transporting complex is trans-
ported via kinesin in squid giant axons. Cell 103,
141–155.
22 Sanchez SA, Chen Y, Mu
¨
ller JD, Gratton E & Hazlett
T (2001) Solution and interface aggregation states of
crotalus atrox venom phospholipase A
2
by two-photon
excitation fluorescence correlation spectroscopy.
Biochemistry 40, 6903–6911.
23 Jankevics H, Prummer M, Izewska P, Pick H, Leufgen
K & Vogel H (2005) Diffusion-time distribution analy-
sis reveals characteristic ligand-dependent interaction
patterns of nuclear receptors in living cells. Biochemistry
44, 11676–11683.
24 Takahashi Y, Okamoto Y, Popiel HA, Fujikake N,
Toda T, Kinjo M & Nagai Y (2007) Detection of poly-
glutamine protein oligomers in cells by fluorescence cor-
relation spectroscopy. J Biol Chem 282, 24039–24048.
25 Kuwajima K (1989) The molten globule state as a clue
for understanding the folding and cooperativity of glob-
ular-protein structure. Proteins 6, 87–103.
26 Maccarrone M, D’Andrea G, Salucci ML, Avigliano L
& Finazzi Agro’ A (1993) Temperature, pH and UV
irradiation effects on ascorbate oxidase. Phytochemistry
32, 795–798.
27 Royer CA (2002) Revisiting volume changes in pres-
sure-induced protein unfolding. Biochim Biophys Acta
1595, 201–209.
28 Di Venere A, Mei G, Gilardi G, Rosato N, De Matteis
F, McKay R, Gratton E & Finazzi Agro
`
A (1998) Res-
olution of the heterogeneous fluorescence in multi-tryp-
tophan proteins: ascorbate oxidase. Eur J Biochem 257,
337–343.
29 Berland KM, So PT & Gratton E (1995) Two-photon
fluorescence correlation spectroscopy: method and
application to the intracellular environment. Biophys J
68, 694–701.
30 Pace CN (1986) Determination and analysis of urea and
guanidine hydrochloride denaturation curves. Methods
Enzymol 131, 266–280.
31 Weber G & Drickamer HG (1983) The effect of high
pressure upon proteins and other biomolecules. Q Rev
Biophys 16, 89–112.
A. Di Venere et al. Monomeric intermediates of AAO
FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS 1593
. Characterization of monomeric substates of ascorbate oxidase Almerinda Di Venere 1,2, *, Eleonora Nicolai 1,2, *, Nicola Rosato 1,2 ,. properties of ascorbate oxidase. FEBS J 273, 5194–5204. 9 Avigliano L & Finazzi-Agro’ A (1997) Biological Func- tion and Enzyme Kinetics of Ascorbate Oxidase. In Multi-copper Oxidase (Messerschmidt. structure of the Monomeric intermediates of AAO A. Di Venere et al. 1592 FEBS Journal 278 (2011) 1585–1593 ª 2011 The Authors Journal compilation ª 2011 FEBS blue oxidase ascorbate oxidase from