Tài liệu Báo cáo khoa học: How disorder influences order and vice versa – mutual effects in fusion proteins containing an intrinsically disordered and a globular protein docx

How disorder influences order and vice versa – mutual
effects in fusion proteins containing an intrinsically
disordered and a globular protein
Ilaria Sambi
1
, Pietro Gatti-Lafranconi
1
*, Sonia Longhi
2
and Marina Lotti
1
1 Dipartimento di Biotecnologie e Bioscienze, Universita
`
di Milano-Bicocca, Italy
2 Architecture et Fonction des Macromole
´
cules Biologiques, Universite
´
Aix-Marseille I et II, France
Introduction
Until very recently, one of the pillars of protein science
has been the so-called structure–function paradigm,
which posits the formation of a unique 3D structure as
the prerequisite for biological function [1]. However,
during the last decade, numerous proteins have been
described that fail to adopt a stable tertiary structure
under physiological conditions and yet display biologi-
cal activity [2]. This condition, defined as intrinsic
disorder, has been found to be widespread in func-
tional proteins. Importantly, disordered regions are
often required for biological activity, indicating that

the lack of stable secondary and tertiary structure is a
Keywords
conformation; fusion proteins; intrinsically
disordered proteins; stability; viral proteins
Correspondence
M. Lotti, Dipartimento di Biotecnologie e
Bioscienze, Universita
`
di Milano-Bicocca,
Piazza della Scienza 2, 20126 Milano, Italy
Fax: +3902 6448 3569
Tel: +3902 6448 3527
E-mail: or
S. Longhi, Architecture et Fonction des
Macromolecules Biologiques (AFMB), UMR
6098 CNRS et Universite
´
s d’Aix-Marseille I
et II, 163, Avenue de Luminy, Case 932,
13288 Marseille, Cedex 09, France
Fax: +33 (0) 4 91 26 67 20
Tel: +33 (0) 4 91 82 55 80
E-mail:
*Present address
Biochemistry Department, University of
Cambridge, UK
(Received 30 June 2010, revised 10 August
2010, accepted 23 August 2010)
doi:10.1111/j.1742-4658.2010.07825.x
Intrinsically disordered proteins (IDPs) are functional proteins either fully

or partly lacking stable secondary and tertiary structure under physiologi-
cal conditions that are involved in important biological functions, such as
regulation and signalling in eukaryotes, prokaryotes and viruses. The func-
tion of many IDPs relies upon interactions with partner proteins, often
accompanied by conformational changes and disorder-to-order transitions
in the unstructured partner. To investigate how disordered and ordered
regions interact when fused to one to another within the same protein, we
covalently linked the green fluorescent protein to three different, well char-
acterized IDPs and analyzed the conformational properties of the fusion
proteins using various biochemical and biophysical approaches. We
observed that the overall structure, compactness and stability of the chime-
ric proteins all differ from what could have been anticipated from the
structural features of their isolated components and that they vary as a
function of the fused IDP.
Abbreviations
GFP, green fluorescent protein; IDP, intrinsically disordered protein.
4438 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
resource rather than a defect. According to this novel
perspective, the straightforward quest for structural
features engaged in a function is moving towards a
dynamic view in which function arises from conforma-
tional freedom. Fully or partly nonstructured proteins
are generally referred to as intrinsically disordered
(IDPs) or intrinsically unstructured proteins, or also as
natively unfolded proteins, a term that emphasizes the
fact that, to fulfil their tasks within the cell, these poly-
peptides rely on existing as a dynamic ensemble of dif-
ferent conformations [3–5]. The intrinsic flexibility of
IDPs indeed provides a clue with respect to their broad
biological functions and high occurrence among pro-

teins with signalling and regulatory roles [4–6]. Consis-
tent with their central position in biological networks,
many disordered proteins are tightly regulated through
the control of their synthesis and degradation and by
post-translational modifications (e.g. phosphorylation)
[7]. Because of its functional relevance, disorder is
widespread in nature, as shown by computational anal-
yses at the genomic level, which indicate that more
than half of all eukaryotic proteins contain unstruc-
tured regions (> 50 residues) and 25–30% of them are
mostly disordered [8]. According to their pivotal role
in signalling and regulation, IDPs are involved in sev-
eral different pathologies [9], such as cancer [10],
as well as cardiovascular [11] and neurodegenerative
diseases [9,12].
Unstructured protein regions often undergo disor-
der-to-order transitions upon interaction with their
partners, as well as upon post-translational modifica-
tions [13,14]. Although the structural effects of inter-
molecular associations have been thoroughly
investigated [15–17], the mutual influence that ordered
and disordered regions exert on each other when they
are embedded in the same protein has received less
attention. This is the case for proteins in which more
compact regions co-exist with fully or mostly unfolded
ones, such as, for example, in KNR4, a 505 residue
yeast protein involved in the coordination of cell wall
synthesis and cell growth [18], the nucleoprotein and
phosphoprotein from measles and Sendai viruses
[19,20] and the Rhabdoviridae phosphoprotein [21].

Although the isolated disordered domains of these lat-
ter proteins have been studied in depth, comprehensive
data on the full-length polypeptides are lacking, with
the data available so far only suggesting that unstruc-
tured regions maintain this feature in the context of
the entire proteins [19–21]. However, evidence that dis-
ordered regions may impact on linked globular
domains arises from work performed in a different
context. In particular, studies by Bae et al. [22] focused
on the prediction of rotational tumbling times of
proteins containing disordered segments, and high-
lighted the effects of the unordered regions on the
properties of covalently linked globular domains (in
this case on the tumbling of the rigid part), with the
extent of the perturbation being proportional to the
length of the disordered region.
With the aim of investigating the reciprocal confor-
mational effect of covalently linked structured and
unstructured protein regions, we fused green fluores-
cent protein (GFP) with disordered fragments of dif-
ferent origin and compactness and investigated the
properties of these fusion proteins using biochemical
and biophysical methods. GFP is a globular protein
with a stable fold and known 3D structure [23]. Its flu-
orophore provides a specific marker to monitor struc-
tural changes in GFP only. As disordered moieties, we
used the unstructured regions of two measles virus
proteins (NTAIL and PNT) and the whole Saccharo-
myces cerevisiae SIC1 protein. Although they are all
IDPs, these proteins have different structural features

and a different extent of disorder. NTAIL is the C-ter-
minal domain (residues 401–525) of the viral nucleo-
protein that is exposed at the surface of the
nucleocapsid [24]. Disorder confers a high structural
plasticity to NTAIL, thus allowing the establishment
of interactions with various partners [25–29]. PNT is
the unstructured N-terminal region of the P protein of
the viral RNA polymerase complex [30,31]. SIC1 is a
284 residue inhibitor of the cyclin-dependent yeast pro-
tein kinase whose conformation in isolation has been
described recently [32,33].
We report that differences in the intrinsic properties
of the IDP (length, a-helix propensity, compactness)
result in fusions with different conformational proper-
ties that are not accounted for by the features of their
components in isolation.
Results and Discussion
Fusion proteins are produced in soluble form
though at levels lower than the constituent
proteins
Plasmids for the expression of fusion proteins were
designed to encode proteins bearing a histidine tag for
immobilized metal-affinity chromatography purifica-
tion. All chimeras consist of the IDP (NTAIL, PNT or
SIC1), a linker of 14 residues containing the TEV pro-
tease cleavage sequence (Glu-Asn-Leu-Tyr-Phe-Gln-
Gly-Ser) and the GFP, in that order (Fig. 1). The
lengths of the resulting fusion proteins are: 386 resi-
dues for NTAIL-GFP (43.2 kDa), 490 residues for
PNT-GFP (53.4 kDa) and 545 residues for SIC1-GFP

I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4439
(61.7 kDa). Isolated IDPs and GFP were expressed
from similar constructs and in the same host cells.
The expression protocol was optimized to obtain
similar amounts of all proteins and to minimize both
the formation of inclusion bodies and spontaneous
proteolysis. Indeed, it has been observed that IDPs are
prone to undergo proteolytic degradation during puri-
fication even upon addition of protease inhibitors to
cell extracts [5]. The culture conditions found to satisfy
all these requirements were: transformed Escherichia
coli BL21 [DE3] cells were grown at 37 °C until D
600
of 0.4–0.5 was reached, then induced with 100 lm of
isopropyl thio-b-d-galactoside at 37 °C for 2 or 6 h,
depending on whether single or fusion proteins were to
be expressed, respectively. Under the above conditions,
all proteins were found to be mainly soluble and prote-
olytic events were negligible (Fig. 2). Despite repeated
attempts (data not shown), we could not improve the
expression level of SIC1-GFP, which systematically
remained very poor. Notably, all the fusion proteins
were fluorescent, thus suggesting that the GFP moiety
adopts a native-like conformation.
Conformational properties of the fusion proteins
vary as a function of the unstructured moiety
NTAIL, PNT and SIC1 have been previously shown
to belong to the family of IDPs on the basis of their
biochemical and biophysical properties [30,32,34].

Accordingly, their far-UV CD spectra recorded at
20 °C show the distinctive IDP profile, characterized
by a large negative peak at 200 nm. The ellipticity val-
ues observed at 200 and 222 nm are consistent with
the existence of some residual helical structure. By con-
trast, the GFP spectrum is typical of a structured pro-
tein with predominant b-strand content, as indicated
by the well-defined positive peak at 195 nm and the
broad negative peak with a minimum at 218 nm
(Fig. 3). Spectra of the fusion proteins combine fea-
tures of ordered and unordered components. Although
minima corresponding to helical structures (a well-
defined inflection point at 203–207 nm and a less pro-
nounced inflection point at 220–222 nm) are clearly
observed, the negative ellipticity values at 200 nm,
together with the low ellipticity in the range 185–
195 nm, suggest the presence of unordered structures
(Fig. 3). Notably, these spectroscopic hallmarks of dis-
order are particularly pronounced for the NTAIL and
SIC1 fusion proteins (Fig. 3A, C), whereas PNT-GFP
exhibits a less disordered nature (Fig. 3B).
To highlight possible mutual effects of disordered
and ordered moieties within the fusion proteins, we
calculated the theoretical average spectra of equimolar
IDP and GFP mixtures by averaging the spectra of the
individual IDP and GFP proteins. Note that each
average spectrum describes what would be expected in
case the two components do not affect each other’s
conformation. We then compared the average spectra
with the experimental measured spectra of equimolar

IDP + GFP mixtures. The CD spectra of NTAIL +
GFP (Fig. 3A) and PNT + GFP (Fig. 3B) mixtures
superimpose quite well onto their respective calculated
theoretical average spectra, indicating that, when the
two separated components are mixed, they do not
undergo any significant structural rearrangement. The
spectra of both NTAIL-GFP and PNT-GFP fusion
proteins are clearly different from those of the mix-
tures either calculated or measured, suggesting that
structural rearrangements are induced in the fusion by
the forced close proximity of the two proteins. In par-
ticular, NTAIL-GFP and PNT-GFP spectra indicate a
lower and higher extent of order with respect to the
average spectra, respectively. By contrast, the spectrum
of the SIC1-GFP fusion protein superimposes onto the
calculated average spectrum, suggesting that the two
domains do not impact on each other’s conformation
Fig. 1. Schematic representation of IDP-GFP constructs. From the
N- to C-terminus, each fusion protein contains the hexahistidine tag
(H
6
), the IDP (NTAIL, PNT or SIC1), a TEV cleavage sequence (TEV)
and the GFP.
Fig. 2. Expression and purification of fused polypeptides and indi-
vidual proteins. M, molecular weight markers; TF, total protein frac-
tion; SOL, soluble protein fraction; proteins purified by immobilized
metal affinity chromatography (IMAC).
Ordered and disordered protein domains I. Sambi et al.
4440 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
when they are covalently linked. More difficult to

explain are the rearrangements observed in the
SIC1 + GFP mixture (Fig. 3C), which suggest that
structural rearrangements only take place when the
two proteins exist as individual moieties in solution.
This observation might be accounted for by the inter-
action depending on orientation factors, with the two
moieties exhibiting a considerably reduced conforma-
tional freedom if covalently linked.
In view of obtaining further insight into the struc-
ture of the fusion proteins, we estimated the content of
a-helices, b-strands, b-turns and unordered regions by
the cdsstr deconvolution method (Fig. 4). Although
this type of analysis does not yield secondary structure
content values that are in perfect agreement with those
derived from structural data obtained by other meth-
ods, it is assumed to be applicable and trustworthy if
its aim is a comparison of the secondary structure con-
tent of a restricted set of spectra obtained under the
same conditions, as in our case [35].
At 20 °C, NTAIL, PNT and SIC1 share a high con-
tent of unordered stretches (50–59%) and a low con-
tent in a-helices and b-strands (2–3% and 22–29%,
respectively), whereas GFP is rich in b-strands (46%)
and exhibits a low content in unordered regions
(20%), in agreement with previous data available for
these proteins [23,30,32,34]. Interesting differences arise
from a comparison of the secondary structure content
of each protein in isolation with that of fusion pro-
teins. We observed that the linkage with GFP does not
alter the b-a-turn-unordered ratio typical of the

unstructured moiety when the IDP is NTAIL or SIC1,
whereas, in the fusion PNT-GFP, the structured part
appears to prevail, raising the percentage of the differ-
ent secondary structures to a value close to that of
GFP alone (Fig. 4). Comparison between the measured
and the averaged secondary structure contents (see
Materials and methods) clearly shows that the second-
ary structure composition of the fusion proteins devi-
ates from the mean of the single contributions (Fig. 5).
Although this analysis does not allow an assessment of
whether the observed deviations in the structural con-
tent reflect structural transitions taking place in only
one of the two moieties or rather reflect structural
rearrangements distributed over the whole polypeptide,
we can speculate that the increase in order in PNT-
GFP likely reflects a gain of structure within PNT.
That PNT possesses an inherent propensity to undergo
a disorder-to-order transition has already been
reported, with this gain of structure concerning the
first 50 residues [30]. Conversely, the less ordered nat-
ure of the NTAIL-GFP fusion protein with respect to
the mean of the secondary structure contents of the
two components could be ascribed either to partial
unfolding of GFP or to loss of residual structure by
NTAIL, with the transiently populated a-helical
regions of the latter [26,34,36–39] adopting preferen-
tially an extended (e.g. disordered) conformation when
linked to GFP.
A
B

C
Fig. 3. Far-UV CD spectra. The CD spectrum of each of the fusion
proteins is compared with that of individual proteins, with the theo-
retical average spectrum and with the spectrum of equimolar pro-
tein mixtures. (A) NTAIL-GFP; (B) PNT-GFP; (C) SIC1-GFP. Spectra
were recorded in 10 m
M sodium phosphate (pH 7.5) at 20 °C.
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4441
Notably, the analysis of the GFP sequence alone
and of GFP bound to any of the three IDPs using the
anchor server, which predicts binding sites within dis-
ordered regions [14,40], shows that the GFP N-termi-
nal part becomes more disordered (Fig. S1). This can
account for the increased disorder measured by far-UV
CD on the NTAIL-GFP fusion protein compared to
the NTAIL + GFP mixture. According to anchor,
NTAIL contains possible binding regions either
(Fig. S1). In this case, however, the experimental data
suggest that these binding regions are not compatible
with GFP binding. anchor predicts numerous possible
binding regions within PNT (Fig. S1). As a result
of order increasing in PNT-GFP compared to PNT +
GFP, one (or more) of the binding regions probably
effectively binds to GFP, albeit not to a great extent,
because this interaction is not measured in the
PNT + GFP mixture. Finally, the predicted SIC1
binding regions (Fig. S1) may interact with GFP, thus
accounting for the increased order in the SIC1 + GFP
mixture compared to the theoretical average. As a

result of steric restrictions, the most likely candidate
for GFP binding is the 183–194 stretch (or the weaker
ones between 214–229 and 253–259) predicted by
anchor.
Regardless of the distribution within the fusion pro-
tein of such folding and unfolding events, we can
clearly state that, in the presence of the same globular
domain (GFP), the overall structure of the fusion pro-
tein varies as a function of the unstructured moiety.
The conformational stability of the chimeras was
investigated by recording variations in the mean resi-
due ellipticity at 195 nm when heating the protein
samples from 20 to 100 °C (Fig. S2). We observed
that, although isolated GFP undergoes a cooperative
unfolding transition between 70 and 90 °C, all IDPs
display almost constant negative mean residue elliptic-
ity values, consistent with the absence of cooperative
unfolding that typifies unstructured proteins, and a
moderate increase of ellipticity at the highest tempera-
tures, consistent with the process of temperature-
induced folding common to several IDPs [41]. Heat
induced transitions recorded for the fusion proteins
were intermediate between these two scenarios. Only
for NTAIL-GFP was a defined transition visible in
the range 75–85 °C range, whereas PNT-GFP and
SIC1-GFP did not display a classical two-state confor-
mational transition. We also monitored the mean resi-
due ellipticity at 195 nm during recooling to 20 °C,
and recorded the CD spectra of the cooled solutions
in the whole range (260–185 nm) to assess the revers-

ibility of unfolding (data not shown). GFP denatur-
ation was found to be only partly reversible, with the
sample exhibiting some helical structure but not the
native b-strand content, whereas the CD spectra of
the IDPs acquired before and after the heating ⁄ recool-
ing process were fully superimposable. Unfolding of
fusion proteins was not fully reversible and showed a
trend very similar to that of isolated GFP, suggesting
that the GFP moiety retains an inability to recover
its native conformation when covalently linked to an
IDP.
Fig. 4. Secondary structure content of IDPs, GFP and fusion pro-
teins. a-helix, b-strands, turns and unordered regions percentages
were calculated using
CDSSTR.
Fig. 5. Deviation from the theoretical secondary structure average
composition. Differences are calculated for each fusion and each
kind of structure by comparing the percentages derived from exper-
imental spectra with the theoretical average compositions as
obtained by averaging the secondary structure content of each indi-
vidual IDP and GFP.
Ordered and disordered protein domains I. Sambi et al.
4442 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
Compactness of fusion proteins depends on the
disordered moiety
In further experiments, we investigated the effect of
the disordered domain on the electrophoretic mobility
of the fusion proteins using SDS ⁄ PAGE migration
analysis (Table 1). IDPs are known to migrate slower
in SDS ⁄ PAGE than globular proteins with the same

molecular mass as a result of their relative enrichment
in acidic residues [5]. The apparent M
r
of the isolated
IDPs as observed in SDS ⁄ PAGE was larger than
expected (Table 1), whereas, for GFP, the expected
and observed values were very close. Because all fusion
proteins exhibited an apparent molecular mass
(M
r App
) significantly higher than expected, we con-
cluded that the presence of a covalently linked IDP is
sufficient to affect GFP migration and that the extent
of this modification is correlated with the specific
disordered component, as suggested by the observed
differences in the M
r App
⁄ M
rth
ratio.
We next addressed the impact of the disordered moi-
ety onto the overall compactness of the fusion proteins
by size exclusion chromatography. Because these studies
are quite demanding in terms of protein amounts, we
only focused onto those proteins (NTAIL-GFP and
PNT-GFP) that could be produced and purified in suffi-
cient quantity (Table 1). In gel filtration experiments,
the elution volume of a given protein can be directly cor-
related with the protein apparent molecular mass by
interpolation with a calibration curve in which elution

volumes of several globular proteins of known size are
correlated with their molecular masses [42]. The hydro-
dynamic radius of a protein (Stokes radius, R
obs
S
) can
then be deduced from its apparent molecular mass and
compared with the theoretical Stokes radius expected
for either the native (R
sN
) or the fully denatured (R
sU
)
form of a protein of the same size (for details, see
Materials and methods). As expected, the GFP hydro-
dynamic behaviour reflected the properties of a globular
protein, with a R
obs
⁄ R
sN
ratio very close to the unit,
whereas ‘aberrant’ elution profiles in gel filtration
experiments were systematically observed for all IDPs.
Previous data showed that NTAIL behaves in gel filtra-
tion as a protein of 36 kDa, whereas its expected mass is
15 kDa [34]. The corresponding R
obs
S
was 27 A
˚

, a value
closer to the radius expected for a fully denatured state
(R
sU
=35A
˚
) compared to globular protein (R
sN
=
19 A
˚
) [34]. In the present study, PNT (expected mass of
25 kDa) was found to elute with an apparent molecular
mass of 115 kDa, in agreement with previous studies
[30]. This very high value of the apparent molecular
mass corresponds to an observed Stokes radius of 41 A
˚
,
which is closer to the value expected for the fully un-
ordered (R
sU
=46A
˚
) than the globular (R
sN
=23A
˚
)
form. The apparent molecular weight of SIC1 was
reported to be 50 kDa instead of 33 kDa, and the

inferred Stokes radius was 30 A
˚
, with the expected R
sU
and R
sN
being 53 and 25 A
˚
, respectively [32].
The apparent molecular masses of both NTAIL-
GFP and PNT-GFP were higher than calculated from
their amino acid sequence ($96 kDa instead of 43 kDa
and $73 kDa instead of 53 kDa, respectively). The
extent of the discrepancy however was not the same.
The calculated R
obs
S
and comparison with the relative
R
sN
and R
sU
showed that the two fusion proteins
have distinctive hydrodynamic behaviours that could
not be anticipated from the characteristics of flexi-
bility of their unstructured component. The R
obs
S
of the
NTAIL-GFP fusion protein ($38 A

˚
) is closer to the
R
sN
(28 A
˚
) than to the R
sU
(61 A
˚
), whereas the R
obs
S
of
PNT-GFP ($35 A
˚
) is closer to the R
sN
(31 A
˚
) than to
the R
sU
(69 A
˚
) (Table 1).
Thus, the hydrodynamic values of NTAIL-GFP and
PNT-GFP proteins do not reflect the sum of the
Table 1. Apparent M
r

of IDPs-GFP, IDPs and GFP derived from SDS ⁄ PAGE and gel filtration. Hydrodynamic radii were inferred from the
apparent molecular mass according to Uversky [46]. ND, not determined.
M
rth
(kDa)
SDS ⁄ PAGE Gel filtration
M
r App
(kDa) M
r App
⁄ M
rth
M
r App
(kDa) M
r App
⁄ M
rth
R
sN
(A
˚
)
R
sU
(A
˚
)
R
obs

S
(A
˚
) R
obs
S
⁄ R
sN
R
obs
S
⁄ R
sU
Reference for gel
filtration
NTAIL-GFP 43 51 1.18 96 2.23 28 61 38 ± 2 1.35 0.62 Present study
PNT-GFP 53 64 1.20 73 1.37 31 69 35 ± 2 1.13 0.51 Present study
SIC1-GFP 62 67 1.08 ND
NTAIL 15 20 1.33 36 2.40 19 35 27 ± 2 1.42 0.77 Longhi et al. [34]
PNT 25 34 1.36 115 4.60 23 46 41 ± 2 1.78 0.89 Karlin et al. [30] and
present study
SIC1 33 39 1.18 50 1.51 25 53 30 ± 2 1.20 0.56 Brocca et al. [32]
GFP 29 30 1.03 38 1.31 25 51 27 ± 2 1.08 0.53 Present study
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4443
behaviour of their single components because isolated
NTAIL is less extended than isolated PNT. The struc-
tural disorder and flexibility typical of isolated NTAIL
is maintained and appears to increase in the fusion,
resulting in a Stokes radius that is even higher than

expected from the mean of the NTAIL and GFP radii.
By contrast, the high flexibility of PNT is not reflected
in PNT-GFP, suggesting that structural modifications
of PNT occur in the fusion protein, in agreement with
the CD data. Because isolated NTAIL is less unor-
dered than isolated PNT but their GFP fusions show
opposite behaviours, these experiments further indicate
that specific, rather than generic, interactions occur.
CD spectra in the near ultraviolet region (250–
350 nm), also known as the aromatic region, reflect the
symmetry of the aromatic amino acid environment
and, consequently, characterize the protein tertiary
structure. Proteins with rigid tertiary structure are typi-
cally characterized by intense near-UV CD spectra,
with unique fine structure, which is reflective of the
unique asymmetric environment of individual aromatic
residues. Conversely, IDPs are characterized by low
intensity near-UV CD spectra with low complexity
[41]. Accordingly, the near-UV CD spectrum of GFP
shows a very pronounced peak at 280 nm, whereas the
spectra of the IDPs are very flat, with no such a clear
peak being detectable (Fig. 6A). Notably, the spectra
of GFP linked to a disordered moiety are much
smoother, with the decrease in the intensity of the
peak being IDP-dependent. Indeed, the spectrum of
the PNT-GFP fusion protein reflects a higher extent of
order than that of GFP fused to NTAIL or to SIC1
(Fig. 6A), in agreement with the data inferred from
both far-UV CD spectroscopy and size exclusion chro-
matography analyses. In the same vein, the visible CD

spectra (Fig. 6B) of GFP alone, as well as of GFP
fusion proteins, show a very pronounced negative peak
at 517 nm, with an intensity in the the order: GFP >
PNT-GFP > NTAIL-GFP > SIC1-GFP. This peak
reflects the asymmetric and therefore rigid environment
of the green chromophore. The gradual reduction in
the intensity of the peak in the fusion proteins is indic-
ative of progressive loss of ordered structure as PNT,
NTAIL or SIC1 are added (Fig. 6B).
In conclusion, near and visible CD data are in good
agreement with the data provided by far-UV and size-
exclusion chromatography studies and, taken together,
they converge to show that PNT-GFP is the most
compact and ordered fusion protein, whereas SIC1-
GFP is the most disordered one.
All the results obtained in the present study so far
point to reciprocal and different effects of the two moie-
ties of the fusion. However, they still do not unravel
whether one of the two domains is more affected in its
conformation; in other words, whether order prevails
on disorder or vice versa. In an attempt to assign these
effects to a specific domain, we analyzed changes in
GFP fluorescence and resistance to proteolysis of the
fusion proteins.
GFP stability is not affected by fusion with the
disordered domain, whereas IDPs are only
marginally protected from proteolysis by the
linked GFP
The presence of a natural chromophore in the globular
part of the fusion provides a sensitive probe for assess-

ing possible conformational changes. The fluorescence
emission spectra of GFP, NTAIL-GFP and PNT-GFP
shared the typical features of the GFP chromophore,
with a well-defined peak at 527 nm and a shoulder at
A
B
Fig. 6. Near-UV and visible CD spectra for tertiary structure analy-
sis. CD spectra acquired in the near-UV (A) and in the visible (B)
wavelength range. Spectra were recorded in 10 m
M sodium phos-
phate (pH 7.5) at 20 °C.
Ordered and disordered protein domains I. Sambi et al.
4444 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
570 nm [43]. The GFP and PNT-GFP emission peaks
at 527 nm were almost superimposable, whereas the
fluorescence intensity of the corresponding peak in
NTAIL-GFP was lower. However, on the basis of the
higher scattering peak of NTAIL-GFP at 475 nm,
such a decrease can likely be attributed to partial pre-
cipitation of the protein rather than to conformational
changes (Fig. 7). When denaturation experiments were
performed, both fusion proteins displayed transitions
in the emitted fluorescence in the temperature range
75–80 °C, which is similar to those observed with the
isolated GFP (data not shown). Such temperature
values are slightly lower than the ones obtained by CD
analysis, as expected for a technique that specifically
targets the active site instead than averaging the whole
protein secondary structure content. The above observa-
tion rules out any effect by the covalently linked IDP on

GFP stability, at least in the protein regions around the
chromophore or critical for its stabilization, and is in
agreement with the results of the spectroscopic analyses
described above, where the GFP moiety, both alone and
IDP-linked, proved unable to recover its native confor-
mation after thermal unfolding.
Globular proteins are rather resistant to proteases,
whereas the extended structure of IDPs makes them
prone to proteolytic attacks [4,5,41,44,45]. For this rea-
son, and in view of understanding which domain is
affected by structural rearrangements, we assessed
whether the presence of GFP is able to protect the
unstructured part of the fusion or, in contrast, GFP
becomes more protease-accessible when linked to an
IDP. In Fig. 8, we show a time-course analysis of a
limited tryptic digestion of NTAIL-GFP and PNT-
GFP, as well as of their isolated IDP moieties. In these
experiments, GFP was very resistant to degradation
even with enzyme : substrate molar ratios as high as
1 : 40 and an incubation time of up to 1 h (data not
shown), whereas both IDPs started to degrade after
1 min of tryptic digestion, and were no longer detect-
able after 5 min (Fig. 8). A significant degradation of
NTAIL was already apparent in the absence of the
enzyme (t
0
in Fig. 8A), consistent with a high prote-
ase-susceptibility of this protein. Interestingly, a
fragment of the same apparent molecular mass
(approximately 17 kDa) was also observed in the sam-

ple containing NTAIL-GFP before the addition of
trypsin (see t
0
in Fig. 8A), suggesting that fusion with
GFP would not protect this specific proteolytic site
within NTAIL. As shown in Fig. 8A, after 20 min of
incubation, full-length NTAIL-GFP disappeared and a
band of the same molecular mass of GFP became
Fig. 7. Fluorescence emission spectra of NTAIL-GFP, PNT-GFP and
GFP at 20 °C. Proteins were excited at 474 nm and spectra were
recorded in the range 465–620 nm in 10 m
M sodium phosphate
(pH 7.5) at 20 °C.
A
B
Fig. 8. Limited proteolyis of NTAIL-GFP, PNT-GFP and their compo-
nent proteins. Two micrograms of purified proteins were incubated
with trypsin at a 1 : 400 enzyme : substrate molar ratio for 1, 2, 5,
10, 20 and 60 min. Samples were separated on 16% SDS ⁄ PAGE
and stained by Coomassie. (A) NTAIL-GFP, NTAIL and GFP. (B)
PNT-GFP, PNT and GFP. M, molecular weight markers; 0¢,
untreated samples. The ‘framed’ bands were further analyzed by
MS (Fig. S3).
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4445
detectable (Fig. 8A). In the course of incubation, a
fragment with an apparent M
r
of 32 kDa (i.e. bigger
than GFP) accumulated and persisted also after 1 h of

digestion (Fig. 8A, box a).
Isolated PNT migrated as a unique band in the con-
trol sample but underwent fast degradation upon tryp-
tic treatment with disappearance of the full-length
polypeptide as early as after 5 min of incubation. Pro-
teolysis of PNT-GFP proceeded with the formation of
a relatively stable fragment with an apparent mass of
31 kDa (Fig. 8B, box b), in addition to that corre-
sponding to GFP alone, already after 1 min of incuba-
tion (Fig. 8B).
Persistent protein fragments (see ‘framed’ bands a
and b in Fig. 8) were processed by tryptic in-gel diges-
tion and the resulting peptides were analyzed by
MS ⁄ MS (Fig. S3). The M
r
, as determined by MS, of
band a from NTAIL-GFP was 31.6 kDa and that of
band b from PNT-GFP was 30.8 kDa. Sequencing
showed that these protease-resistant fragments encom-
pass the trypsin cutting sites at position 128 in
NTAIL-GFP and at positions 226, 235 and 236 of
PNT-GFP, respectively. These results indicate that the
complete proteolytic digestion of both IDPs requires
longer incubation when they are fused with GFP, with
a proteolytic fragment still containing part of the IDP
being detectable after as long as 60 min of incubation
in both cases (Fig. 8). That the GFP sensitivity
towards proteolysis was not affected by its linkage to
an unstructured part was checked in two additional
experiments. Western blotting analysis of a time-

course digestion (up to 20 min) of NTAIL-GFP and
PNT-GFP ruled out the presence of fragments react-
ing with anti-GFP antibodies smaller than full-length
GFP (Fig. S4). Moreover, the fragment of approxi-
mately 20 kDa produced from PNT-GFP after 20 min
of tryptic digestion (Fig. 8B, band c) was found to
span the same amino acid sequence as the band of the
same size that was detectable after 2 min of digestion
of PNT alone (Fig. 8B, band d). This protein
fragment encompasses a region of PNT upstream that
is in band b (Fig. S3), thus ruling out the possibility
that it could correspond to a GFP proteolytic frag-
ment. On the basis of these two lines of experimental
evidence, we concluded that the proteolytic resistance
of GFP is not affected by the presence of the fused
IDP.
In both fusion proteins, proteolysis occurs within
the unstructured part only and proceeds from the
N-terminus towards the GFP moiety, resulting in the
generation of a trypsin-resistant fragment that contains
GFP after 1 h of incubation. The persistence of GFP
in these fragments, besides highlighting its resistance
towards proteolysis, also suggests protection of the
C-terminal region of the IDP by the GFP moiety.
However, we cannot exclude the possibility that pro-
tection only arises from steric hindrance (i.e. from a
reduced substrate accessibility to trypsin), rather than
being the result of local structural rearrangement
within the IDP.
How do ordered and disordered parts affect each

other?
In conclusion, we have observed that different IDPs
fused to the same globular protein result in polypep-
tides with distinctive secondary structure content and
compactness that are not merely the average of their
two components. The finding that their overall struc-
ture and compactness are not consistent with those
that are predicted on the basis of the behaviour of the
isolated IDP was intriguing. Indeed, although PNT
alone is more flexible than NTAIL, PNT-GFP is by
far more structured and compact that the NTAIL
fusion. This observation may suggest that linkage with
GFP confers the two IDPs with folding propensities
that differ from those of the isolated NTAIL and PNT
proteins. Nonetheless, our attempts to highlight spe-
cific structural rearrangements within either the IDP or
the GFP moiety that could account for the specific
conformational features observed were hindered by the
complex nature of proteins. Association with a disor-
dered moiety left GFP almost unchanged, whereas the
IDP was marginally stabilized towards proteolysis.
However, despite the higher compactness of the PNT-
GFP fusion protein with respect to NTAIL-GFP, no
higher proteolytic resistance of PNT-GFP could be
detected. It could be speculated that the high flexibility
of IDPs prevents the formation of stable interactions,
causing delocalized structural rearrangements that
failed to manifest in the experiments conducted in the
present study.
Materials and methods

Construction of expression plasmids
The NTAIL-GFP, PNT-GFP and SIC1-GFP constructs
were obtained in two steps: the single IDP-encoding
sequences were cloned in the pET22 plasmid (Novagen,
Madison, WI, USA) and then the GFP encoding sequence
was inserted downstream. The cloning strategy was differ-
ent in each case and is described below.
The DNA fragment encoding NTAIL with a hexahisti-
dine tag fused to its N-terminus was obtained by PCR from
the pDest14 ⁄ N
TAILHN
plasmid [36]. To remove the NcoI
Ordered and disordered protein domains I. Sambi et al.
4446 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
restriction site at position 321, amplification was carried
out in two separated reactions yielding products N
1
(from
nucleotides 1–330) and N
2
(from nucleotides 331–396). N
1
was amplified using a forward primer (FWN
1
:5¢-TACCG
TTAACATCGATATGCATCATCATCATCATCATAC-3¢),
designed to introduce a ClaI restriction site at nucleotide
position –6 and a reverse primer (REVN
1
:5¢-CCTGCC

ATTGCTTGCAGCC-3¢) that introduced a silent mutation
at nucleotide position 321 resulting in the suppression of
the NcoI site. Product N
2
was amplified with a forward pri-
mer (FWN
2
:5¢-GGCTGCAAGCAATGGCAGG-3¢) that
introduced the same silent mutation as above at nucleotide
position 321 and a reverse primer (REVN
2
:5¢-ATCGCC
ATGGTCCCGGGCATATGGGATCCCTGGAAGTACA
GGTTTTCGTCTAGAAGATTTCTGTC-3¢) designed to
remove the NTAIL stop codon and to introduce a fragment
encoding a TEV cleavage sequence and a NcoI restriction
site at position +38 after the end of the NTAIL sequence.
N
1
and N
2
were mixed, digested with DpnI to remove the
methylated DNA template, and used as the template in a
PCR reaction with primers FWN
1
and REVN
2
to yield the
complete NTAIL amplification product.
The DNA fragment encoding PNT with an N-terminal

hexahistidine tag was obtained by PCR using the
pET21a ⁄ PNT-
H6
plasmid [30] as the template. The forward
primer (5¢-TACCGTTAACATCGATATGCATCATCATC
ATCATCATGC-3¢) was designed to insert a ClaI restric-
tion site at nucleotide position )6, whereas the reverse
primer (5¢ -ATCGCCATGGTCCCGGGCATATGGGATC
CCTGGAAGTACAGGTTTTCCTTTTTAATGGGTGTC
CC-3¢) was design ed to remove the stop codon and to
introduce a DNA fragment encoding a TEV cleavage
sequence and a NcoI restriction site at position +38 after
the end of the PNT sequence.
The DNA sequence encoding SIC1 with a hexahistidine
tag fused to its N-terminus was obtained by PCR from the
plasmid pET21 ⁄ SIC1 [32] with a forward primer (5¢-TAC
CTGGCCAATGAATATGCATCATCATCATCATCATA
CTCCGTCGACCCCACC-3¢) designed to introduce a
hexahistidine tag and a ClaI restriction site at nucleotide
position –6 and a reverse primer (5 ¢-A TCGCCATGGTC
CCGGGCATATGGGATCCCTGGAAGTACAGGTTTT
CGCCATGCTCTTGATCCC-3¢) designed to remove the
stop codon and to introduce a DNA fragment encoding a
TEV cleavage sequence and a NcoI restriction site at posi-
tion +38 after the end of the SIC1 gene fragment.
PCR reactions contained 2.5 mm dNTPs, 5 lm of each
primer, 10 ng of plasmid DNA and 5 U of Triple MasterÔ
DNA Polymerase (Eppendorf, Hamburg, Germany). The
amplification program was: after a first denaturation step
at 94 °C for 5 min, 25 cycles of 20 s at 94 °C, 20 s at 50 ° C

and 2 min at 72 °C were performed, followed by a final
elongation step of 10 min at 72 °C.
All PCR products (NTAIL, PNT and SIC1) were
digested with DpnI and purified by precipitation with
ethanol, restricted with ClaI and NcoI, checked by agarose
(0.8%, w ⁄ v) gel electrophoresis and purified from the gel
(QIAquick Gel Extraction Kit; Qiagen, Valencia, CA,
USA). The pET22 vector was digested with NdeI, filled-in
with the Klenow fragment of the E. coli DNA polymerase I
(New England Biolabs, Beverly, MA, USA) to produce
blunt ends, and finally cleaved with NcoI. In this way, the
sequence pelB, allowing targeting to the periplasm of pro-
teins expressed from pET22, was removed. The digested
PCR products and pET22 were ligated with T4 Ligase
(New England Biolabs). The final constructs are referred to
as pET ⁄ NTAIL, pET ⁄ PNT and pET ⁄ SIC1.
The GFP gene (cloned from a pET19b ⁄ GFP plasmid)
was then inserted downstream NTAIL, PNT and SIC1 at
the NcoI and ScaI sites to obtain pET ⁄ NTAIL-GFP,
pET ⁄ PNT-GFP and pET ⁄ SIC1-GFP. These constructs
encode for fusion proteins bearing a 14 residues linker
containing the TEV cleavage sequence between the two
components. The final constructs were transformed into
the E. coli DH5a strain (Novagen) and the sequence of
their ORFs was checked by DNA sequencing on both
strands.
Expression and purification of fusion proteins
The E. coli BL21[DE3] strain (Novagen) was used as the
host for heterologous expression. Transformed cells were
grown overnight at 37 °C in low-salt LB medium contain-

ing 100 mgÆL
)1
ampicillin, diluted 1 : 50 in 200 mL of the
same broth and incubated at 37 °C until until D
600
of 0.4–
0.5 was reached. Induction was performed by adding
100 lm isopropyl thio-b-d-galactoside. Cells were then
grown at 37 ° C for either 2 h when expressing single IDPs
or for 6 h when expressing fusions and GFP.
Cells were collected by centrifugation and resuspended
in 2 mL of lysis buffer (50 mm sodium phosphate, pH 8.0,
300 mm NaCl, 5 mm imidazole) containing the protease
inhibitors cocktail P8465 (Sigma-Aldrich, St Louis, MO,
USA). After 20 min of incubation on ice, cells were dis-
rupted by sonication (four cycles of 10 s each at 50%
power output). Cell extracts were centrifuged for 30 min
at 10 000 g at 4 °C and the His-tagged proteins recovered
as soluble proteins from the supernatant. They were then
purified by immobilized metal-affinity chromatography on
aNi
2+
-nitrilotriacetic acid resin (Qiagen). The clarified
lysate was added to a pre-packed resin suspension (2 mL
of resin per 200 mL of culture), eluted by gravity and
then reloaded four or five times on the column. After
washing with 50 mm sodium phosphate (pH 8.0), 300 mm
NaCl buffer containing increasing concentrations of imid-
azole (25–50 mm), proteins were eluted with 50 mm
sodium phosphate (pH 8.0), 300 mm NaCl and 250 mm

imidazole. When required, buffer exchange was performed
by gel filtration on PD-10 columns (GE Healthcare, Mil-
waukee, WI, USA) and samples were concentrated with a
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4447
Microcon
Ò
Ultra-15 centrifugal filter device (molecular
mass cut-off, 10 000 Da) (Millipore Corp., Billerica, MA,
USA).
Protein fractions were analyzed on 12% polyacrylamide
gels stained by GelCode Blue (Pierce, Rockford, IL, USA).
Proteins concentration was determined with the Bradford
assay using BSA as the standard.
CD
The CD spectra of proteins (0.1–0.2 mgÆmL
)1
for far-UV
measurements and 1 mgÆmL
)1
for near-UV and visible mea-
surements) in 10 mm sodium phosphate buffer (pH 7.5)
were recorded on a J-815 spectropolarimeter (Jasco Corp.,
Easton, MD, USA), using either a 1 mm or 1 cm path-
length quartz cuvette for far-UV or for near-UV and visible
domains, respectively. A Peltier thermoregulation system
was used when recording the spectra. All the experiments
were performed in triplicate. Denaturation ⁄ renaturation
spectra were obtained measuring the CD signal at a fixed
wavelength (195 nm) when progressively heating from 20 to

100 °C, and then recooling to 20 °C. This wavelength was
chosen because it allows variation in b-strand content to be
monitored. Measurements were performed with a data pitch
of 0.1 °C and a temperature slope of 5 °CÆmin
)1
. A com-
plete CD spectrum at 20 °C was also acquired after and
before the heating ⁄ recooling process. Spectra at 20 °C were
measured in the range 185–260 nm for far-UV measures,
260–400 nm for near-UV measures and 400–750 nm
for UV-visible measures, with 0.2 nm data pitch and
20 nmÆmin
)1
scanning speed. All spectra were corrected for
buffer contribution, averaged from three independent
acquisitions and smoothed by using a third-order least
square polynomial fit.
Mean ellipticity values per residue [h] were calculated as:
½h¼
3300mDA
lcn
where DA represents the difference in the adsorption
between circularly polarized right and left light of the
protein corrected for blank, m is the protein molecular
mass in daltons, l is the path length (0.1 or 1 cm), c is
the protein concentration (mgÆmL
)1
) and n is the number
of residues.
The theoretical average ellipticity values per residue,

[h]
Ave
, expected for a protein mixture in which no second-
ary structure rearrangements take place upon mixing
equimolar amounts of protein 1 and protein 2 were calcu-
lated as:
½h
Ave
¼
fð½h
1
Á n
1
Þþð½h
2
Á n
2
Þg
ðn
1
þ n
2
Þ
where [h]
1
and [h]
2
correspond to the measured mean ellip-
ticity values per residue of proteins 1 and 2, respectively,
and n

1
and n
2
correspond to the number of residues of
proteins 1 and 2.
When recording spectra of equimolar (4 lm each) protein
mixtures, [h] were calculated as:
½h¼
3300 DA
l
c
1
n
1
m
1
Þþ
c
2
n
2
m
2
Þ
ohn
where l is the path length (0.1 or 1 cm), n
1
or n
2
represent

the number of residues, m
1
or m
2
is the molecular mass
(Da), and c
1
or c
2
is the protein concentration (mgÆmL
)1
)
for proteins 1 and 2, respectively.
The experimental data were then analyzed using the
cdsstr software ( />home.shtml) using SDP42 as the reference dataset. The
cdsstr deconvolution method was used to estimate the
content of a-helices, b-strands, b-turns and unordered
regions of individual and fusion proteins. The secondary
structure composition expected for fusion proteins with no
reciprocal structural impact was obtained by averaging the
content of a-helices, b-strands, b-turns and unordered
regions of the individual intrinsically unstructured proteins
and GFP moieties.
Analytical gel filtration and calculation of Stokes
radii
Gel filtration was performed on an A
¨
KTA purifier liquid-
chromatography system (GE Healthcare), using a pre-
packed, 30 · 1 cm SuperdexTM 75 HR column (GE

Healthcare). Chromatography was carried out at room
temperature in 50 mm sodium phosphate (pH 8.0), 200 mm
NaCl at a flow rate of 0.5 mLÆmin
)1
. Protein elution was
monitored by checking absorbance at 280 nm. Apparent
molecular masses of proteins eluted from the column were
deduced from a calibration curve obtained by loading
200 lL of the following standards (2 mgÆmL
)1
): aprotinin
(6500 Da), horse cytochrome c (12 400 Da), horse myoglo-
bin (17 600 Da), chicken ovalbumin (45 000 Da), apoferr-
itin (80 000 Da), alcohol dehydrogenase (150 000 Da) and
BSA (66 400 Da) (Sigma-Aldrich).
The theoretical hydrodynamic radii (Stokes radius) for a
native (R
sN
) or fully unfolded (R
sU
) protein were calculated
according to Uversky [46]:
log R
sN
¼ð0:369  log M
r
ÞÀ0:254 ð1Þ
log R
sU
¼ð0:533  log M

r
ÞÀ0:682 ð2Þ
where M
r
is the molecular mass (as inferred from the amino
acid sequence). M
r
was calculated using the protparam
tool at the expasy server ( The
experimentally observed Stokes radii (R
obs
S
) were deduced
by inserting the apparent M
r
(as observed in gel filtration
experiments) in Eqn (1).
Ordered and disordered protein domains I. Sambi et al.
4448 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
Fluorescence spectroscopy
Fluorescence emission of the GFP chromophore was mea-
sured by using a Cary Eclipse (Varian Inc., Palo Alto, CA,
USA) spectropolarimeter using 1 · 1 cm quartz cuvette
containing 1 lm protein in 10 mm sodium phosphate (pH
7.5). For spectra at 20 °C, the excitation wavelength was
474 nm and the emission spectra were recorded in the range
465–620 nm. Using the same excitation wavelength
(474 nm), thermal denaturation studies were performed by
recording fluorescence emission at 527 nm when heating
samples from 20 to 95 °C, and recooling back to 20 °C.

Limited proteolysis
Proteolysis of purified proteins (4 lm) was performed in
10 mm sodium phosphate buffer, pH 7.5, at room tempera-
ture with Trypsin (Promega Corp., Madison, WI, USA).
The enzyme : substrate molar ratio was 1 : 400. After 1, 2,
5, 10, 20 and 60 min of incubation, aliquots from the reac-
tion mix were collected and immediately mixed with
SDS ⁄ PAGE loading buffer and boiled to stop the reaction.
The extent of proteolysis was assayed by 16% SDS ⁄ PAGE.
In-gel digestion and mass spectrometry
Bands of interest were excised from the polyacrylamide
gels, cut into small pieces and destained by repeated wash-
ing cycles alternating 50 mm ammonium hydrogen carbon-
ate and pure acetonitrile. After complete destaining, gel
particles were dehydrated by acetonitrile, covered with a
trypsin solution (12.5 ngÆmL
)1
in 50 mm ammonium hydro-
gen carbonate, pH 8.0) and incubated for 1 h on ice. Once
the excess of liquid was removed, gel pieces were covered
with a solution of 50 mm ammonium hydrogen carbonate
(pH 8.0) and incubated overnight at 37 °C.
Tryptic peptides were extracted from gel particles
by repeated cycles of incubation in pure acetonitrile or
1% formic acid. Samples were lyophilized, resuspended
in 1% formic acid, and desalted by ZipTip (Millipore,
Carrigtwohill, Ireland) before ESI-MS ⁄ MS analysis.
ESI-MS experiments were performed on a hybrid Quad-
rupole-Time-of-Flight mass spectrometer (Q-Star Elite;
Applied Biosystems, Foster City, CA, USA) equipped with

a nano-electrospray ionization sample source. Metal-coated
borosilicate capillaries (Proxeon, Odesnse, Denmark) with a
medium-length emitter tip of 1 lm internal diameter were
used. The instrument was calibrated with a standard solu-
tion of renin (Applied Biosystems). Spectra of tryptic pep-
tides were acquired in the range 500–1500 m ⁄ z, at room
temperature, with an accumulation time of 1 s, ion-spray
voltage of 1300 V and declustering potential of 60 V, with
application of active information-dependent acquisition,
using rolling collision energy to fragment peptides for
MS ⁄ MS analysis. Peptide identification was performed
using mascot software (Matrix Science Ltd, London, UK)
with the parameters: two trypsin missed cleavages; peptide
tolerance, 0.6 Da; MS ⁄ MS tolerance, 0.6 Da; peptide
charges, 2+ and 3+. Only monoisotopic masses were con-
sidered as precursor ions.
Acknowledgements
I.S. acknowledges financial support from the EXTRA
programme of UNIMIB-Cariplo, allowing her to carry
out part of this work in Marseille. The authors wish to
thank Antonino Natalello and Silvia Maria Doglia for
their assistance with the fluorescence spectroscopy, as
well as for critically reading the manuscript. We are
indebted to Maria Samalikova for performing the mass
spectrometry. We also wish to thank the anonymous
reviewer who suggested carrying out near-UV and visi-
ble CD experiments. This work was supported by a
grant from the University Milano-Bicocca (Fondo di
Ateneo per la Ricerca) to M.L. and by a grant
from the Agence Nationale de la Recherche, specific

program ‘Microbiologie et Immunologie’, ANR-05-
MIIM-035-02, to S.L.
References
1 Anfinsen CB (1973) Principles that govern the folding
of protein chains. Science 181, 223–230.
2 Uversky VN & Dunker AK (2010) Understanding
protein non-folding. Biochim Biophys Acta 1804,
1231–1264.
3 Wright PE & Dyson HJ (1999) Intrinsically unstruc-
tured proteins: re-assessing the protein structure-
function paradigm. J Mol Biol 293, 321–331.
4 Dunker AK, Lawson JD, Brown CJ, Williams RM,
Romero P, Oh JS, Oldfield CJ, Campen AM, Ratliff
CM, Hipps KW et al. (2001) Intrinsically disordered
protein. J Mol Graph Model 19, 26–59.
5 Tompa P (2002) Intrinsically unstructured proteins.
Trends Biochem Sci 27, 527–533.
6 Dunker AK, Brown CJ, Lawson JD, Iakoucheva LM &
Obradovic Z (2002) Intrinsic disorder and protein
function. Biochemistry 41, 6573–6582.
7 Gsponer J, Futschik ME, Teichmann SA & Babu MM
(2008) Tight regulation of unstructured proteins: from
transcript synthesis to protein degradation. Science 322,
1365–1368.
8 Oldfield CJ, Cheng Y, Cortese MS, Brown CJ, Uversky
VN & Dunker AK (2005) Comparing and combining
predictors of mostly disordered proteins. Biochemistry
44, 1989–2000.
9 Tompa P (2009) Structural disorder in amyloid fibrils:
its implication in dynamic interactions of proteins.

FEBS J 276, 5406–5415.
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4449
10 Iakoucheva LM, Brown CJ, Lawson JD, Obradovic Z
& Dunker AK (2002) Intrinsic disorder in cell-signaling
and cancer-associated proteins. J Mol Biol 323, 573–584.
11 Cheng Y, LeGall T, Oldfield CJ, Dunker AK &
Uversky VN (2006) Abundance of intrinsic disorder in
protein associated with cardiovascular disease. Biochem-
istry 45, 10448–10460.
12 Chiti F & Dobson CM (2006) Protein misfolding, func-
tional amyloid, and human disease. Annu Rev Biochem
75, 333–366.
13 Dyson HJ & Wright PE (2005) Intrinsically unstruc-
tured proteins and their functions. Nat Rev Mol Cell
Biol 6, 197–208.
14 Meszaros B, Simon I & Dosztanyi Z (2009) Prediction
of protein binding regions in disordered proteins. PLoS
Comput Biol 5, e1000376.
15 Radhakrishnan I, Perez-Alvarado GC, Parker D,
Dyson HJ, Montminy MR & Wright PE (1997) Solu-
tion structure of the KIX domain of CBP bound to the
transactivation domain of CREB: a model for
activator:coactivator interactions. Cell 91, 741–752.
16 Oldfield CJ, Meng J, Yang JY, Yang MQ, Uversky VN
& Dunker AK (2008) Flexible nets: disorder and
induced fit in the associations of p53 and 14-3-3 with
their partners. BMC Genomics 9(Suppl 1), S1.
17 Uversky VN, Oldfield CJ, Midic U, Xie H, Xue B,
Vucetic S, Iakoucheva LM, Obradovic Z & Dunker AK

(2009) Unfoldomics of human diseases: linking protein
intrinsic disorder with diseases. BMC Genomics
10(Suppl 1), S7.
18 Durand F, Dagkessamanskaia A, Martin-Yken H,
Graille M, Van TilbeurghH, Uversky VN & Francois
JM (2008) Structure-function analysis of Knr4 ⁄ Smi1, a
newly member of intrinsically disordered proteins
family, indispensable in the absence of a functional
PKC1-SLT2 pathway in Saccharomyces cerevisiae.
Yeast 25, 563–576.
19 Longhi S & Oglesbee M (2010) Structural disorder
within the measles virus nucleoprotein and phosphopro-
tein. Protein Pept Lett 17, 961–978.
20 Jensen MR, Bernado P, Houben K, Blanchard L, Mar-
ion D, Ruigrok RW & Blackledge M (2010) Structural
disorder within sendai virus nucleoprotein and
phosphoprotein: insight into the structural basis of
molecular recognition. Protein Pept Lett 17, 952–960.
21 Gerard FC, Ribeiro EdeA Jr, Leyrat C, Ivanov I,
Blondel D, Longhi S, Ruigrok RW & Jamin M (2009)
Modular organization of rabies virus phosphoprotein.
J Mol Biol 388, 978–996.
22 Bae SH, Dyson HJ & Wright PE (2009) Prediction of
the rotational tumbling time for proteins with disor-
dered segments. J Am Chem Soc 131, 6814–6821.
23 Ormo M, Cubitt AB, Kallio K, Gross LA, Tsien RY &
Remington SJ (1996) Crystal structure of the Aequorea
victoria green fluorescent protein. Science 273, 1392–1395.
24 Heggeness MH, Scheid A & Choppin PW (1980)
Conformation of the helical nucleocapsids of paramyx-

oviruses and vesicular stomatitis virus: reversible coiling
and uncoiling induced by changes in salt concentration.
Proc Natl Acad Sci USA 77, 2631–2635.
25 Bourhis JM, Johansson K, Receveur-Brechot V,
Oldfield CJ, Dunker KA, Canard B & Longhi S (2004)
The C-terminal domain of measles virus nucleoprotein
belongs to the class of intrinsically disordered proteins
that fold upon binding to their physiological partner.
Virus Res 99, 157–167.
26 Gely S, Lowry DF, Bernard C, Jensen MR, Blackledge
M, Costanzo S, Bourhis JM, Darbon H, Daughdrill G
& Longhi S (2010) Solution structure of the C-terminal
X domain of the measles virus phosphoprotein and
interaction with the intrinsically disordered C-terminal
domain of the nucleoprotein. J Mol Recognit 23, 435–
447.
27 Zhang X, Bourhis JM, Longhi S, Carsillo T, Buccellato
M, Morin B, Canard B & Oglesbee M (2005) Hsp72
recognizes a P binding motif in the measles virus N
protein C-terminus. Virology 337, 162–174.
28 Couturier M, Buccellato M, Costanzo S, Bourhis JM,
Shu Y, Nicaise M, Desmadril M, Flaudrops C, Longhi
S & Oglesbee M (2010) High affinity binding between
Hsp70 and the C-terminal domain of the measles virus
nucleoprotein requires an Hsp40 co-chaperone. J Mol
Recognit 23, 301–315.
29 Iwasaki M, Takeda M, Shirogane Y, Nakatsu Y,
Nakamura T & Yanagi Y (2009) The matrix protein
of measles virus regulates viral RNA synthesis and
assembly by interacting with the nucleocapsid protein.

J Virol 83, 10374–10383.
30 Karlin D, Longhi S, Receveur V & Canard B (2002)
The N-terminal domain of the phosphoprotein of
Morbilliviruses belongs to the natively unfolded class
of proteins. Virology 296, 251–262.
31 Karlin D, Ferron F, Canard B & Longhi S (2003)
Structural disorder and modular organization in
Paramyxovirinae N and P. J Gen Virol 84, 3239–3252.
32 Brocca S, Samalikova M, Uversky VN, Lotti M, Vanoni
M, Alberghina L & Grandori R (2009) Order propensity
of an intrinsically disordered protein, the cyclin-depen-
dent-kinase inhibitor Sic1. Proteins 76, 731–746.
33 Mittag T, Orlicky S, Choy WY, Tang X, Lin H, Sicheri
F, Kay LE, Tyers M & Forman-Kay JD (2008)
Dynamic equilibrium engagement of a polyvalent ligand
with a single-site receptor. Proc Natl Acad Sci USA
105, 17772–17777.
34 Longhi S, Receveur-Brechot V, Karlin D, Johansson K,
Darbon H, Bhella D, Yeo R, Finet S & Canard B
(2003) The C-terminal domain of the measles virus
nucleoprotein is intrinsically disordered and folds upon
binding to the C-terminal moiety of the phosphopro-
tein. J Biol Chem 278, 18638–18648.
Ordered and disordered protein domains I. Sambi et al.
4450 FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS
35 Greenfield NJ (2006) Using circular dichroism spectra
to estimate protein secondary structure. Nat Protoc 1,
2876–2890.
36 Bourhis JM, Receveur-Brechot V, Oglesbee M, Zhang
X, Buccellato M, Darbon H, Canard B, Finet S &

Longhi S (2005) The intrinsically disordered C-terminal
domain of the measles virus nucleoprotein interacts with
the C-terminal domain of the phosphoprotein via two
distinct sites and remains predominantly unfolded.
Protein Sci 14, 1975–1992.
37 Morin B, Bourhis JM, Belle V, Woudstra M, Carriere F,
Guigliarelli B, Fournel A & Longhi S (2006) Assessing
induced folding of an intrinsically disordered protein by
site-directed spin-labeling electron paramagnetic reso-
nance spectroscopy. J Phys Chem B 110, 20596–20608.
38 Belle V, Rouger S, Costanzo S, Liquiere E, Strancar J,
Guigliarelli B, Fournel A & Longhi S (2008) Mapping
alpha-helical induced folding within the intrinsically dis-
ordered C-terminal domain of the measles virus nucleo-
protein by site-directed spin-labeling EPR spectroscopy.
Proteins 73, 973–988.
39 Kavalenka A, Urbancic I, Belle V, Rouger S, Costanzo
S, Kure S, Fournel A, Longhi S, Guigliarelli B & Stran-
car J (2010) Conformational analysis of the partially
disordered measles virus N(TAIL)-XD complex by
SDSL EPR spectroscopy. Biophys J 98, 1055–1064.
40 Dosztanyi Z, Meszaros B & Simon I (2009) ANCHOR:
web server for predicting protein binding regions in
disordered proteins. Bioinformatics 25, 2745–2746.
41 Uversky VN (2002) Natively unfolded proteins: a point
where biology waits for physics. Protein Sci 11, 739–756.
42 Receveur-Brechot V, Bourhis JM, Uversky VN, Canard
B & Longhi S (2006) Assessing protein disorder and
induced folding. Proteins 62, 24–45.
43 Chalfie M, Tu Y, Euskirchen G, Ward WW & Prasher

DC (1994) Green fluorescent protein as a marker for
gene expression. Science 263, 802–805.
44 Hubbard SJ (1998) The structural aspects of limited
proteolysis of native proteins. Biochim Biophys Acta
1382, 191–206.
45 Fontana A, Zambonin M, Polverino deLauretoP, De
FilippisV, Clementi A & Scaramella E (1997) Probing
the conformational state of apomyoglobin by limited
proteolysis. J Mol Biol 266, 223–230.
46 Uversky VN (1993) Use of fast protein size-exclusion
liquid chromatography to study the unfolding of
proteins which denature through the molten globule.
Biochemistry 32, 13288–13298.
Supporting information
The following supplementary material is available:
Fig. S1. Analysis of the amino acid sequences of the
IDPs of GFP and of the three fusion proteins using
the anchor server.
Fig. S2. Heat-induced unfolding.
Fig. S3. Sequence identification of tryptic fragments by
MS ⁄ MS analysis.
Fig. S4. NTAIL-GFP and PNT-GFP fragmentation
with trypsin.
This supplementary material can be found in the
online version of this article.
Please note: As a service to our authors and readers,
this journal provides supporting information supplied
by the authors. Such materials are peer-reviewed and
may be re-organized for online delivery, but are not
copy-edited or typeset. Technical support issues arising

from supporting information (other than missing files)
should be addressed to the authors.
I. Sambi et al. Ordered and disordered protein domains
FEBS Journal 277 (2010) 4438–4451 ª 2010 The Authors Journal compilation ª 2010 FEBS 4451