G Model

BC-4693; No. of Pages 8 ARTICLE IN PRESS

The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Contents lists available at ScienceDirect

The International Journal of Biochemistry

& Cell Biology
journal homepage: www.elsevier.com/locate/biocel

Personalized liposome–protein corona in the blood of breast, gastric

and pancreatic cancer patients
Valentina Colapicchioni a,1 , Martina Tilio b,1 , Luca Digiacomo b,c , Valentina Gambini b ,
Sara Palchetti c , Cristina Marchini b , Daniela Pozzi c,d , Sergio Occhipinti e,f , Augusto Amici b ,
Giulio Caracciolo c,∗
a
Istituto Italiano di Tecnologia, Center for Life Nano Science@Sapienza, Viale Regina Elena 291, 00161 Rome, Italy
b
School of Biosciences and Veterinary Medicine, University of Camerino, Via Gentile III da Varano, 62032 Camerino (MC), Italy
c
Department of Molecular Medicine, Sapienza University of Rome, Viale Regina Elena 291, 00161 Rome, Italy
d
Istituti Fisioterapici Ospitalieri, Istituto Regina Elena, Via Elio Chianesi 53, 00144 Rome, Italy
e
Departments of Molecular Biotechnology and Health Sciences, University of Torino, Torino, Italy
f
Center for Experimental Research and Medical Studies (CERMS), AOU Città della Salute e della Scienza di Torino, Torino, Italy

a r t i c l e i n f o a b s t r a c t

Article history: When nanoparticles (NPs) are dispersed in a biofluid, they are covered by a protein corona the composition
Received 3 August 2015 of which strongly depends on the protein source. Recent studies demonstrated that the type of disease
Received in revised form 7 September 2015 has a crucial role in the protein composition of the NP corona with relevant implications on personalized
Accepted 8 September 2015
medicine. Proteomic variations frequently occur in cancer with the consequence that the bio-identity of
Available online xxx
NPs in the blood of cancer patients may differ from that acquired after administration to healthy vol-
unteers. In this study we investigated the correlation between alterations of plasma proteins in breast,
Keywords:
gastric and pancreatic cancer and the biological identity of clinically approved AmBisome-like liposomes
Protein corona
Liposome
as determined by a combination of dynamic light scattering, zeta potential analysis, one-dimensional
Breast cancer sodium dodecyl sulfate polyacrylamide gel electrophoresis (1D-SDS-PAGE) and semi-quantitative densit-
Gastric cancer ometry. While size of liposome–protein complexes was not significantly different between cancer groups,
Pancreatic cancer the hard corona from pancreatic cancer patients was significantly less negatively charged. Of note, the
hard corona from pancreatic cancer patients was more enriched than those of other cancer types this
enrichment being most likely due to IgA and IgG with possible correlations with the autoantibodies pro-
ductions in cancer. Given the strict relationship between tumor antigen-specific autoantibodies and early
cancer detection, our results could be the basis for the development of novel nanoparticle-corona-based
screening tests of cancer.
© 2015 Elsevier Ltd. All rights reserved.

1. Introduction
Due to the peculiar properties that arise when a material is
reduced to the nanoscale, nanoparticles (NPs) are being utilized
in almost limitless applications in biomedical research (Hafner
et al., 2014). The hardest challenge occurs when NPs are dis-
persed in a biofluid, because of their tendency to interact with
biomolecules, such as proteins, lipids and sugars driven either by
a potential energy gradient, or just by diffusion. These molecules
instantaneously adsorb onto the NPs’ surface creating a complex
biomolecular coating around the particles, referred to as the NP
∗ Corresponding author.
E-mail address: giulio.caracciolo@uniroma1.it (G. Caracciolo).
1
These authors contributed equally to this work.
corona (Mahon et al., 2012; Monopoli et al., 2011a). What is clear
to date is that the new identity given by the corona is the main
factor controlling biodistribution, therapeutic effect and nanotox-
icity of NPs in the body (Walkey and Chan, 2012). Exposure time
has been identified as a key factor shaping the NP biomolecular
corona (Dell’Orco et al., 2014; Tenzer et al., 2013). The biomolecular
corona is highly dynamic in nature with its composition depending
both on the physicochemical properties of the particles (Caracciolo
et al., 2010; Docter et al., 2014; Lundqvist et al., 2008; Mahmoudi
et al., 2011, 2012; Monopoli et al., 2011b; Röcker et al., 2009; Zhang
et al., 2011) (e.g. material, size, surface charge, shape) and the char-
acteristics of biological media including protein source (Caracciolo
et al., 2014b), concentration (Caracciolo et al., 2011; Monopoli et al.,
2011b) and temperature (Mahmoudi et al., 2013a). All the studies
reported so far used prevalently plasma from healthy volunteers,
while very few earlier studies have characterized the interaction

http://dx.doi.org/10.1016/j.biocel.2015.09.002
1357-2725/© 2015 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
2 V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx
of nanoscale particles with plasma collected from patients with
different clinical manifestations (Caracciolo et al., 2014a; Hajipour
et al., 2014). Recently, Hajipour et al. (2014) demonstrated that the
type of disease critically affects the corona composition of com-
mercially available silica and polystyrene NPs. This observation
let the authors introduce the innovative concept of personalized
protein corona (PPC). The most direct implication of the PPC is
that precise corona information is needed for personalized ther-
apeutics and diagnostic (Hamad-Schifferli, 2015). This might have
obvious impact in cancer where hundreds of proteins are differ-
entially expressed (Sallam, 2015). Proteomic variations in cancer
imply that the bio-identity of NPs in the blood of cancer patients
may differ from that acquired after administration to healthy
volunteers. Nonetheless, since each cancer type can specifically
alter the levels of relevant proteins, it is reasonable to assume
that patients with specific cancers may have specific nanoparti-
cle coronas. As such, the main goal of this investigation was to
determine whether the pathologically alterations of plasma pro-
teins in cancer influenced the protein corona and dependent NP
features. To test this suggestion, we incubated liposomes with
plasma from humans with histologically proven breast, gastric
and pancreatic cancer with similar severity. The selected cancer
types are respectively the second, fifth and twelfth most common
cancers for both sexes (www.wcrf.org). Liposomes are currently
tested for numerous applications in nano-biomedicine and are
already used in cancer treatment (Allen and Cullis, 2013; Barenholz,
2012; Caracciolo, 2015). In particular, a lipid composition made
of hydrogenated soy phosphatidylcholine (HSPC), 1,2-distearoyl-
sn-glycero-3-[phospho-rac-(1-glycerol)] (DSPG), and Cholesterol
(Chol) that constitutes the basis of the liposomal amphotericin B
agent AmBisome (Adler-Moore and Proffitt, 2002) was employed
in this study. Since AmBisome liposomes are anionic, as those used
in anti-cancer therapy (Caracciolo, 2015), they are excellent model
systems to study the interaction between plasma proteins and
clinically approved liposomes. Basically, AmBisome is indicated
in the treatment of severe systemic and/or deep mycoses and in
the empirical treatment of presumed fungal infections. Anyway,
patients with cancer who receive chemotherapy have an increased
chance of getting fungal infections (Johansen and Gøtzsche, 2000).
Antifungal drugs are therefore often given prophylactically to can-
cer patients, or when they have a fever. To understand whether
human plasma (HP) collected from breast, gastric and plasma
cancer patients could alter the corona structure and composition
here we used a combination of one-dimensional sodium dodecyl
sulfate polyacrylamide gel electrophoresis (1D-SDS-PAGE), semi-
quantitative densitometry, dynamic light scattering (DLS) and zeta
potential analysis. Of note, exposure to plasma of pancreatic can-
cer significantly altered the structure and composition of the
liposome–protein corona.
2. Materials and methods
2.1. Liposomes preparation
Hydrogenated soy phosphatidylcholine (HSPC), 1,2-distearoyl-
sn-glycero-3-[phospho-rac-(1-glycerol)] (sodium salt) (DSPG),
cholesterol (Chol.) were purchased from Avanti Polar Lipids
(Alabaster, AL). AmBisome-like liposomes were synthesized
according to (Marchini et al., 2011) with these molar ratios:
HSPC:DSPG:cholesterol (2.5:1:1.2) (Adler-Moore and Proffitt,
2002). Lipid films were hydrated with phosphate saline buffer
(PBS) 10 mmol l−1 (pH 7.4) to a final lipid concentration 1 mg/ml.
The obtained liposome suspension was extruded 20 times through
a 0.05 ␮m polycarbonate carbonate filter by the Avanti Mini-
Extruder (Avanti Polar Lipids, Alabaster, AL).
Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
2.2. Protein corona isolation
Milestone studies (Caracciolo et al., 2011; Cedervall et al., 2007;
Lundqvist et al., 2008; Monopoli et al., 2011b) have shown that
the aggregation state of NPs, thickness and composition of the pro-
tein corona can change with HP concentration. However, saturation
occurs at around 40–80% plasma concentration depending on the
NP’s nature. According to this indication, we have previously inves-
tigated the effect of plasma concentration on the composition of the
liposome–protein corona (Caracciolo et al., 2011). Saturation was
found to occur at 50%, thus here we used 50% plasma concentration
(i.e., 1:1 liposome:HP volume ratio). For proteomics experiments,
100 ␮l liposome suspension was mixed with HP (collected from
both healthy and cancer patients) (1:1 v/v) and incubated for 60 min
at 37 ◦ C. After incubation, protein–liposomes corona complexes
were isolated by centrifugation for 15 min at 14,000 rpm. Then
pellets were washed three times with 200 ␮l of PBS to remove
unbound proteins obtaining the “hard corona.”
2.3. Human specimens
HP samples were isolated by centrifugation from heparinized
venous blood of cancer patients not previously treated with radio or
chemotherapy. Cancer patients included patients with breast can-
cer (n = 10), gastric cancer (n = 10) or pancreatic adenocarcinoma
(n = 10) with similar severity, recruited at the Centro Oncologico
Ematologico Subalpino (COES), AOU Città della Salute e della
Scienza di Torino (Torino, Italy) with informed consent. Blood sam-
ples were immediately processed after drawing and plasma were
stored at −80 ◦ C until use. Human blood from healthy volunteers
was obtained by venipuncture and prepared at the Department of
Experimental Medicine (Sapienza University of Rome) according
to the institutional bioethics code; a K2-EDTA anticoagulant and
protease inhibitor cocktail were immediately added. Blood sam-
ples were immediately processed after drawing and plasma were
stored at −80 ◦ C until use. When used, aliquots were thawed at 4 ◦ C
and then let to warm at room temperature
2.4. Ethics statement
The human studies were conducted according to the Declara-
tion of Helsinki principles. Human investigations were performed
after approval of the study by the Scientific Ethics Committee of
AOU Città della Salute e della Scienza di Torino, Torino, Italy (Prot.
No. 0012068). Written informed consent was received from each
participant prior to inclusion in the study and specimens were de-
identified prior to analysis.
2.5. Size and zeta experiments
For size and zeta-potential experiments, liposome–protein
complexes were prepared by incubating 10 ␮l of each lipid disper-
sion, pre-diluted with 490 ␮l of a PBS:H2 O solution (1:80 v/v), with
10 ␮l of HP pre-diluted with 490 ␮l of a PBS:H2 O solution (1:80
v/v). Dilution did not modify the pH of HP (pH = 7.45 ± 0.03). After
1-h incubation, samples were ready to be analyzed. Size and zeta-
potential measurements were performed on a Zetasizer Nano ZS90
(Malvern, UK) at room temperature. Dynamic light scattering (DLS)
allowed to retrieve normalized intensity auto-correlation functions
that were analyzed by the CONTIN method, which allows one to
obtain the distribution of the diffusion coefficient D of the parti-
cles. This coefficient is converted into an effective hydrodynamic
radius RH by using the Stokes–Einstein relationship RH = KBT/(6D),
where KBT is the thermal energy and the solvent viscosity. Polydis-
persity index (PDI) of DLS measurements can be found in Table
S1 of the Supplementary Material. The electrophoretic mobility
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx
1 2 3 4 5 6 7 8 9 10 Average
A 400
300
DH (nm)
200
100
0
B Healthy Breast cancer Gastric cancer Pancreatic cancer
0 the formation of a rich protein corona whose thickness is given by
Zeta-potenial (mV)
-5
-10
-15
-20
Fig. 1. (A) Hydrodynamic diameter of liposomes following 1-hour incubation with
human plasma (HP) of healthy and cancer patients. Collected samples were not
pooled and each liposome-HP sample was analyzed. Values are mean ± s.d. from
three independent experiments. (B) Zeta-potential of liposomes after 1-hour expo-
sure to human plasma (HP) collected from healthy donors and cancer patients.
Values are mean ± s.d. from three independent experiments. The significance of size
and zeta-potential measurements has been statistically evaluated by Student’s t-test
and is reported in Table S1.
measurements were performed by means of the laser Doppler
electrophoresis technique, the same apparatus used for size mea-
surements. From the mobility u one can calculate the zeta-potential
using the Smoluchowski relation zeta-potential = u/ε, where u and ε
are the viscosity and the permittivity of the solvent phase, respec-
tively. All the experiments were carried out in triplicate. Results
are given as average ± standard deviation. For both size and zeta-
potential measurements, HP samples were not pooled, but each
sample was let to interact with liposomes.
2.6. 1D-SDS-PAGE experiments
The hard corona-coated NPs were re-suspended in 20 ␮l of
Laemmli Loading buffer 1X (10 ␮l ␤ME/1 ml Laemmli Loading
buffer) and boiled for 5 min at 100 ◦ C. Identical volumes (10 ␮l) of
each sample were loaded on a gradient polyacrylamide gel (4–20%
Criterion TGX precast gels, Bio-Rad) and run at 100 V for approxi-
mately 150 min. Gels were rinsed in double-distilled water (ddH2 O)
and fixed over night for 12 h in a de-staining solution (MeOH 50%
in ddH2 O) with gentle agitation at room temperature, with at least
one solution change. In order to determine the differences in corona
composition in patients affected by different types of cancer, we
stained the proteins with highly sensitive silver-ammonia solu-
tion for 30 min. Subsequently, gels were rinsed 3 times in ddH2 O,
40 s each rinse, then incubated with acidic developer solution from
3 to 5 min. Once all the protein bands started to clearly appear,
developer solution was removed and stop solution (acetic acid 10%,
methanol 45% in ddH2 O) was added. Pictures of gels were captured
by KODAK Digital DC120.
3. Results
3.1. Size results
Synthesized liposomes were small in size (DH = 131.6 ± 4.2 nm).
In Fig. 1 panel A, we report DLS data for liposome–protein com-
plexes after 1-h incubation with HP from healthy donors and
cancer patients (documents of liposome size characterization are
reported in Table S2). As evident, the hydrodynamic diameter of
Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
3
liposomes increased by roughly 80–100 nm (210 nm < DH < 230 nm)
with respect to bare liposomes. Size increase of lipid–protein com-
plexes is due to the corona formation on single nanoparticles
and to the agglomeration of distinct nanoparticle-corona units
(Caracciolo et al., 2013). A recently proposed spherical packing
model (M’Hallah et al., 2013) predicts that the smallest sphere
packing two stiff spheres of radius R has radius 2R. On average,
size of liposome–protein complexes was smaller than double the
diameter of liposomes (≈260 nm). As a result, incubation with HP
did not produce appreciable liposome clustering, but resulted in
dPC = 1/2[DH (t = 1 h) − DH (t = 0)] ≈ 40–50 nm. This value is in agree-
ment with those previously reported for other liposome–protein
complexes (Caracciolo et al., 2014b). First, we observe that the aver-
age size of liposome–protein complexes did not appreciably differ
among the healthy group (DH = 211.8 ± 9.5 nm). On the other side,
liposome–protein complexes from cancer patients were more het-
erogeneous in size compared to the control. In particular, for each
cancer type, a single size value compatible with liposome aggrega-
tion was found (breast cancer: patient #1, DH = 298 ± 30 nm; gastric
cancer: patient #1, DH = 333 ± 75 nm; pancreatic cancer: patient #2,
DH = 131.6 ± 4.2 nm). However, the hydrodynamic size from DLS
did not give evidence of significantly different changes between
the groups (statistical significance is reported in Table S1).
3.2. Zeta-potential results
In Fig. 1 panel B, we report the zeta-potential of
liposome–protein complexes after interaction with HP collected
from both healthy volunteers and cancer patients (documents of
liposome zeta-potential characterization are reported in Table S3).
As Fig. 1 panel B shows, zeta-potential of liposome–protein com-
plexes was more positive (−14.1 mV < zeta potential < −12.7 mV)
than that of bare liposomes (zeta-potential = −33.6 ± 1.4 mV).
This charge alteration is likely due to the absorption of positively
charged plasma proteins. Few differences were generally observed
among healthy volunteers (zeta-potential = −13.7 ± 1.2 mV). Like-
wise, zeta potential values of the hard coronas from breast and
gastric cancer were similar to each other. On the opposite, the
liposome–protein corona of pancreatic cancer patients was signif-
icantly less negative compared to those of breast cancer patients,
gastric patients and healthy subjects (statistical significance is
reported in Table S1).
3.3. 1D SDS-PAGE results
After incubation with the various HP samples, the hard coronas
of liposomes were analyzed under the same experimental con-
ditions. Fig. 2 panel A shows 1D SDS/PAGE gel results in which
liposomes were incubated with HP from five healthy volunteers.
Some general observations could be made: (i) the bands of the
hard coronas from various healthy cases are very similar to each
other, but they are not always the same; (ii) the intensities of pro-
tein bands in the hard coronas of the same subject are similar; (iii)
the band intensities of the hard coronas from various healthy cases
vary. Collectively, these findings suggest that differences in the hard
coronas of various healthy cases exist. For instance, some additional
protein bands seem to be present only in lanes of healthy patient 1,
while one protein band disappeared only in lanes of healthy patient
5. These are caused by personalized plasma alterations that occur
among persons due to general health characteristics. To compare
the hard coronas from liposomes incubated with different plasma
samples a semi-quantitative densitometry analysis based on the
protein band intensity was performed. First, we provided insight
into the total protein content of hard coronas by calculating the
total lane intensity of proteins recovered from liposome–protein
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
4 V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Fig. 2. (A) 1D SDS-PAGE gel of human plasma (HP) proteins obtained from liposome–protein complexes following incubation with 50% human plasma (HP) from various
healthy donors. (B) Histograms representing the total lane intensity of proteins recovered from liposomes incubated with HP collected from various healthy donors. (C)
Histogram of the alterations in band intensity. Proteins are grouped by molecular weight. (D) The relative densitometry results of four major bands from the gel in panel (A).
To normalize the data the ladder total intensity was fixed to one. The axis ranges have been fixed for direct comparison with the figures in Supplementary Information.

complexes. Fig. 2 panel B shows that the amounts of protein in dif-
ferent hard coronas obtained from the same plasma were almost
the same, while the total protein content from various healthy cases
varies. For example, the total band intensity of healthy patient 1
was found to be almost two-fold that of patient 5. Fig. 2 panel C
shows classification of band intensities based on four molecular
weight ranges (i.e., >100 kDa, 60–100 kDa, 30–60 kDa and <30 kDa).
As evident, relevant differences in each molecular weight range
were found. More in detail, we observed the presence of four major
bands located at ≈110 kDa, 90 kDa, 75 kDa and 37 kDa (Fig. 2, panel

Fig. 3. (A) One-dimensional SDS-PAGE gel of human plasma proteins obtained from liposome–protein complexes following incubation with human plasma (50%) from
various breast cancer patients. (B) Histograms representing the total lane intensity of proteins recovered from liposomes incubated with human plasma (50%) collected from
various breast cancer patients. (C) Histogram of the alterations in band intensity during incubation of liposomes with plasma (50%) from various breast cancer patients.
Proteins are grouped by molecular weight. (D) The relative densitometry results of four major bands from the gel in panel (A). To normalize the data, the ladder total intensity
was fixed to one.

Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx 5

Fig. 4. (A) One-dimensional SDS-PAGE gel of human plasma proteins obtained from liposome–protein complexes following incubation with human plasma (50%) from
various gastric cancer patients. (B) Histograms representing the total lane intensity of proteins recovered from liposomes incubated with human plasma (50%) collected from
various gastric cancer patients. (C) Histogram of the alterations in band intensity during incubation of liposomes with plasma (50%) from various gastric cancer patients.
Proteins are grouped by molecular weight. (D) The relative densitometry results of four major bands from the gel in panel (A). To normalize the data, the ladder total intensity
was fixed to one.

D). Based on the intensities of selected bands, it is clear that marked
differences among individuals existed likely due to personalized
plasma alterations. The same approach was used to analyze the pro-
tein patterns of the various hard coronas of breast (Fig. 3), gastric
(Fig. 4) and pancreatic cancer patients (Fig. 5). As found for healthy
individuals, it was clear that the bands of the hard coronas from
various cases were the same, but their intensities were very differ-
ent from each other. In Fig. 6, we show the total lane intensities for
the hard coronas of breast, gastric and pancreatic cancer patients
compared to the control. We observe that the total lane intensity
was definitely higher for pancreatic cancer. According to results of
Fig. 7, the intensities of the four selected bands (≈110 kDa, 90 kDa,
75 kDa and 37 kDa) were much more intense in the corona of pan-
creatic cancer patients. Because the signal intensity is proportional
to the protein adsorption, this result indicates that the hard coronas
of pancreatic cancer patients are significantly more enriched than
those of other cases.
4. Discussion
Bio–nano interactions can be defined as the interactions
between nanoscale and biological systems. In the last decade,
researchers have deeply investigated the bio–nano interactions
between NPs and biological fluids (Caracciolo et al., 2011; Cedervall
et al., 2007; Del Pino et al., 2014; Mahmoudi et al., 2011, 2012,
2013b; Monopoli et al., 2011a; Prapainop et al., 2012; Salvati
et al., 2013; Walkey et al., 2012). Even though a NP itself is
the easiest example of a homogeneous sphere of one material,
interaction with biomolecules present in biological fluids such as
sugars, lipids and proteins, will lead to alteration of the NP sur-
face resulting in the rapid formation of a complex coating, referred
to as protein corona, whose composition is largely shaped by
the physical-chemical properties of the NPs. The protein corona
is at the outmost part of the NP thus it is the interface seen
by the physiological environment and determines the biological
identity of the NP. The complexity of the biological environ-
ment can deeply modify the composition of the protein corona
with possible biological implications (Caracciolo et al., 2014a,
2014b; Hajipour et al., 2014). While some dependence has been
understood, whether altered expressions of proteins, as those
occurring in the plasma of cancer patients, may have an impact
on the biological identity of nanoparticles remain to be resolved.
Although exact causes are not clearly defined, hundreds of plasma
proteins are differentially expressed, whether increase or decrease,
as a cause or consequence of cancer (Sallam, 2015). As a result,
the bio-identity of NPs might be affected by protein alteration
in the blood of cancer patients. Furthermore, each malignancy
can alter the levels of specific proteins with possible implica-
tions on the nanoparticle coronas. The main aim of this study
was to investigate whether incubation of liposomes with HP sam-
ples from various cancer types may lead to significant changes
in the nanoparticle-corona. To test this suggestion, we incubated
AmBisome-like liposomes with plasma from humans with his-
tologically proven breast, gastric and pancreatic cancer that are
respectively the second, fifth and twelfth most common cancers
for both sexes and have been all associated to relevant pro-
tein alterations. On the other side, AmBisome-like liposomes are
excellent model systems of many approved anti-cancer liposo-
mal drugs (Allen and Cullis, 2013). As above mentioned, the hard
corona is formed rapidly (during the first seconds of incubation) and
its composition slightly changes with time (Barrán-Berdón et al.,
2013). In this study, we used the most common method of hard
corona preparation, i.e., incubation of liposome with plasma for 1 h.
First, we performed a systematic study aimed at understanding if
size and charge of liposome–protein complexes were significantly
different between healthy and cancer groups. Incubation with HP

Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
6 V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx

Fig. 5. (A) One-dimensional SDS-PAGE gel of human plasma proteins obtained from liposome–protein complexes following incubation with human plasma (50%) from
various pancreatic cancer patients. (B) Histograms representing the total lane intensity of proteins recovered from liposomes incubated with human plasma (50%) collected
from various pancreatic cancer patients. (C) Histogram of the alterations in band intensity during incubation of liposomes with plasma (50%) from various pancreatic cancer
patients. Proteins are grouped by molecular weight. (D) The relative densitometry results of four major bands from the gel in panel (A). To normalize the data, the ladder
total intensity was fixed to one.

1 2 3 4 5 6 7 8 9 10 Average of cancer. Our findings may be relevant in the context of anti-cancer

9
Normalized Total Intensity (a.u.)

personalized medicine, i.e., the attempt to tailor medicinal treat-

8
ments (e.g., chemotherapeutic doses) to the specific characteristics
7 of each individual patient (Cole and Holland, 2015). In the next
6 future, personalized medicine may aid in reducing undesirable side
5 effects of medications and theoretically produce better global out-
4
comes not only in cancer but also in the majority of diseases. In
a pioneering study (Hajipour et al., 2015), Hajipour and cowork-
3
ers showed that the protein corona of graphene oxide (GO) sheets
2 strongly depends on the type of disease/condition of humans from
1 which the plasma is collected. Of note, Hajipour and coworkers
0 also demonstrated that such PPCs do trigger different biological
Healthy Breast cancer Gastric cancer Pancreatic cancer responses. Those findings represent a hallmark of the biological
Fig. 6. Histograms representing the total lane intensity of proteins recovered from
impact of PPC and shall have a profound impact on the guidance
liposomes incubated with 50% human plasma collected from healthy donors and for the development of new generations patient-specific nano bio-
cancer patients. For each subject, protein patterns were acquired in triplicate. materials for clinical applications. Anyway, surface charges of the
protein decorations from breast and gastric cancer were very sim-
ilar to that of healthy subjects, while the PPC of pancreatic cancer
did not produce appreciable liposome aggregation, but resulted in patients was significantly less negative than those of breast cancer
the formation of a thick protein corona. Notably, liposome–protein patients, gastric cancer patients and healthy subjects (calculation
complexes from cancer patients were more heterogeneous in size of statistical significance can be found in Table S1). This finding
compared to the control suggesting that the composition of the indicates that the hard corona from pancreatic cancer is mostly
hard corona varies among patients with the same diseases and with enriched of cationic plasma proteins. 1D SDS-PAGE gels strongly
similar severity. In particular, for each cancer type, a single size approved the significant changes in the hard corona of pancre-
value compatible with liposome aggregation was found. However, atic cancer. The total lane intensity was assessed to estimate the
as shown in Fig. 1 panel A, the hydrodynamic size from DLS did not relative amount of protein present in the fraction derived from
give evidence of significantly different changes between the groups. each patient. As Fig. 3 clearly shows, the hard corona from pan-
Zeta-potential of liposome–protein complexes was about 20 mV creatic cancer patients was definitely more enriched than those
more positive than that of bare liposomes (Fig. 1 panel B). This of other cancer types. Such enrichment was prevalently due to
finding means that positively charged plasma proteins are highly two bands located at ≈37 kDa and 75 kDa (Fig. 4). Even though a
enriched in the corona of AmBisome-like liposomes. As found for precise identification of corona proteins goes beyond the scope
the size of liposome–protein complexes, the zeta-potential of the of this investigation, some considerations can be made. In many
hard corona was found to vary among patients with the same type previous investigations we applied nanoliquid chromatography

Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx 7

A 1 2 3 4 5 6 7 8 9 10 Average B 2.5 1 2 3 4 5 6 7 8 9 10 Average

2.5

Normalized Total Intensity (a.u.)

Normalized Total Intensity (a.u.)
110 kDa 90 kDa
2 2

1.5 1.5

1 1

0.5 0.5

0 0
Healthy Breast cancer Gastric cancer Pancreatic cancer Healthy Breast cancer Gastric cancer Pancreatic cancer

C 2.5
1 2 3 4 5 6 7 8 9 10 Average D 2.5 1 2 3 4 5 6 7 8 9 10 Average
Normalized Total Intensity (a.u.)

Normalized Total Intensity (a.u.)

75 kDa 37 kDa
2 2

1.5 1.5

1 1

0.5 0.5

0 0
Healthy Breast cancer Gastric cancer Pancreatic cancer Healthy Breast cancer Gastric cancer Pancreatic cancer

Fig. 7. The relative densitometry results of four major bands obtained from the gels reported in Figs. 2–5.

tandem mass spectrometry to identify and quantify plasma pro- less negative. According to the recently introduced concept of PPC
teins that compose the liposome–protein corona (Caracciolo et al., (Hajipour et al., 2014), we showed that hard coronas vary among
2015). Around 75 kDa, we found some complement components, healthy subjects as well as for identical diseases. Of note, the hard
serum albumin (69 kDa), Vitamin K (75 kDa) and serotransferrin corona from pancreatic cancer patients was more enriched than
(77 kDa). On the other side, the band at 37 kDa is mainly enriched those of other cancer types. This enrichment was most likely due to
of heavy chains of immunoglobulins alpha (IgA) and immunoglobu- IgA and IgG with possible correlations with the autoantibodies pro-
lins gamma (IgG). Of note, this suggestion is in very good agreement ductions in cancer. Since tumor autoantibodies appear definitely
with a previous study (Caracciolo et al., 2014a) showing that plasma earlier before diagnosis of cancer, and autoantibodies have been
samples collected from pancreatic cancer patients exhibit signif- discovered in several cancer types, our results could be the basis for
icant change in concentration of Igs. The increased amount of the development of nanoparticle-corona-based technologies aimed
human IgG found in the protein corona does correlate with the at screening for the early detection of cancer.
autoantibodies production in cancer patients as part of the immun-
odefense against tumor (Zheng et al., 2015). This is note worthy Acknowledgements
since tumor antigen-specific autoantibodies appear from months
to years before diagnosis of cancer, and autoantibodies have been GC and DP acknowledge support by the Italian Minister for
discovered in several cancer types. Nonetheless, the nanoparticle- University and Research (“Futuro in Ricerca 2008”, Grant No.
corona technology can work as a “nano-concentrator,” since it RBFR08TLPO) and by the Italian Minister of Health (“Progetto Gio-
allows detecting increased amount of human IgG in the protein vani Ricercatori 2011-2012”, Grant No. GR-2011-02350094). This
corona formed in cancer patients’ sera even when the total quan- work has been supported by Fondazione Veronesi that granted S.O.
tity of human IgG in the blood sera is about the same for cancer
and noncancer group (Zheng et al., 2015). The sequestration of pro-
Appendix A. Supplementary data
teins around the NP surface provides an excellent opportunity to
probe these proteins that are present in the biological milieu and
Supplementary data associated with this article can be found,
responsible for tumorigenesis. These proteins are actually detected
in the online version, at http://dx.doi.org/10.1016/j.biocel.2015.09.
because of their enrichment on the NP surface, which indicates
002.
the relevance of this technology to identify proteins that would
otherwise not have been revealed due to low abundance. Hence,
References
we believe that liposome surfaces provide a unique platform to
sequester, enrich, and identify proteins for cancer detection. Adler-Moore, J., Proffitt, R.T., 2002. AmBisome: liposomal formulation, structure,
mechanism of action and pre-clinical experience. J. Antimicrob. Chemother.
49, 21–30.
Allen, T.M., Cullis, P.R., 2013. Liposomal drug delivery systems: from concept to
5. Conclusion
clinical applications. Adv. Drug Deliv. Rev. 65, 36–48.
Barenholz, Y.C., 2012. Doxil® —the first fda-approved nano-drug: lessons learned. J.
In this study we investigated the correlation between alter- Control. Release 160, 117–134.
Barrán-Berdón, A.L., Pozzi, D., Caracciolo, G., Capriotti, A.L., Caruso, G., Cavaliere, C.,
ations of plasma proteins in breast, gastric and pancreatic cancer
et al., 2013. Time evolution of nanoparticle-protein corona in human plasma:
and the biological identity of AmBisome-like liposomes. Size of relevance for targeted drug delivery. Langmuir 29, 6485–6494.
liposome–protein corona was not significantly different between Caracciolo, G., 2015. Liposome–protein corona in a physiological environment:
healthy and cancer groups. Zeta potential values of the hard coro- challenges and opportunities for targeted delivery of nanomedicines.
Nanomed. Nanotechnol. Biol. Med. 11, 543–557.
nas from breast and gastric cancer were similar to each other, while Caracciolo, G., Callipo, L., De Sanctis, S.C., Cavaliere, C., Pozzi, D., Laganà, A., 2010.
that of pancreatic cancer patients was found to be significantly Surface adsorption of protein corona controls the cell internalization

Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
G Model
BC-4693; No. of Pages 8 ARTICLE IN PRESS
8 V. Colapicchioni et al. / The International Journal of Biochemistry & Cell Biology xxx (2015) xxx–xxx
mechanism of DC-Chol-DOPE/DNA lipoplexes in serum. Biochim. Biophys. Acta
1798, 536–543.
Caracciolo, G., Caputo, D., Pozzi, D., Colapicchioni, V., Coppola, R., 2014a. Size and
charge of nanoparticles following incubation with human plasma of healthy
and pancreatic cancer patients. Colloids Surf. B 123, 673–678.
Caracciolo, G., Cardarelli, F., Pozzi, D., Salomone, F., Maccari, G., Bardi, G., et al.,
2013. Selective targeting capability acquired with a protein corona adsorbed
on the surface of 1, 2-dioleoyl-3-trimethylammonium propane/DNA
nanoparticles. ACS Appl. Mater. Interfaces 5, 13171–13179.
Caracciolo, G., Pozzi, D., Capriotti, A.L., Cavaliere, C., Foglia, P., Amenitsch, H., et al.,
2011. Evolution of the protein corona of lipid gene vectors as a function of
plasma concentration. Langmuir 27, 15048–15053.
Caracciolo, G., Pozzi, D., Capriotti, A.L., Cavaliere, C., Piovesana, S., Amenitsch, H.,
et al., 2015. Lipid composition: a “key factor” for the rational manipulation of
the liposome–protein corona by liposome design. RSC Adv. 5,
5967–5975.
Caracciolo, G., Pozzi, D., Capriotti, A.L., Cavaliere, C., Piovesana, S., La Barbera, G.,
et al., 2014b. The liposome–protein corona in mice and humans and its
implications for in vivo delivery. J. Mater. Chem. B 2, 7419–7428.
Cedervall, T., Lynch, I., Foy, M., Berggård, T., Donnelly, S.C., Cagney, G., et al., 2007.
Detailed identification of plasma proteins adsorbed on copolymer
nanoparticles. Angew. Chem. Int. Ed. 46, 5754–5756.
Cole, J.T., Holland, N.B., 2015. Multifunctional nanoparticles for use in theranostic
applications. Drug Deliv. Transl. Res. 5, 295–309.
Del Pino, P., Pelaz, B., Zhang, Q., Maffre, P., Nienhaus, G.U., Parak, W.J., 2014. Protein
corona formation around nanoparticles – from the past to the future. Mater.
Horiz. 1, 301–313.
Dell’Orco, D., Lundqvist, M., Linse, S., Cedervall, T., 2014. Mathematical modeling of
the protein corona: implications for nanoparticulate delivery systems.
Nanomedicine 9, 851–858.
Docter, D., Bantz, C., Westmeier, D., Galla, H.J., Wang, Q., Kirkpatrick, J.C., et al.,
2014. The protein corona protects against size-and dose-dependent toxicity of
amorphous silica nanoparticles. Beilstein J. Nanotechnol. 5,
1380–1392.
Hafner, A., Lovrić, J., Lakoš, G.P., Pepić, I., 2014. Nanotherapeutics in the EU: an
overview on current state and future directions. Int. J. Nanomed. 9, 1005.
Hajipour, M.J., Laurent, S., Aghaie, A., Rezaee, F., Mahmoudi, M., 2014. Personalized
protein coronas: a “key” factor at the nanobiointerface. Biomater. Sci. 2,
1210–1221.
Hajipour, M.J., Raheb, J., Akhavan, O., Arjmand, S., Mashinchian, O., Rahman, M.,
et al., 2015. Personalized disease-specific protein corona influences the
therapeutic impact of graphene oxide. Nanoscale 7, 8978–8994.
Hamad-Schifferli, K., 2015. Exploiting the novel properties of protein coronas:
emerging applications in nanomedicine. Nanomedicine 10, 1663–1674.
Johansen, H.K., Gøtzsche, P.C., 2000. Amphotericin B lipid soluble formulations
versus amphotericin B in cancer patients with neutropenia. Cochrane Database
Syst. Rev.
Lundqvist, M., Stigler, J., Elia, G., Lynch, I., Cedervall, T., Dawson, K.A., 2008.
Nanoparticle size and surface properties determine the protein corona with
possible implications for biological impacts. Proc. Natl. Acad. Sci. 105,
14265–14270.
M’Hallah, R., Alkandari, A., Mladenovic, N., 2013. Packing unit spheres into the
smallest sphere using VNS and NLP. Comput. Oper. Res. 40, 603–615.
Please cite this article in press as: Colapicchioni, V., et al., Personalized liposome–protein corona in the blood of breast, gastric and
pancreatic cancer patients. Int J Biochem Cell Biol (2015), http://dx.doi.org/10.1016/j.biocel.2015.09.002
Mahmoudi, M., Abdelmonem, A.M., Behzadi, S., Clement, J.H., Dutz, S., Ejtehadi,
M.R., et al., 2013a. Temperature: the “ignored” factor at the NanoBio interface.
ACS Nano 7, 6555–6562.
Mahmoudi, M., Laurent, S., Shokrgozar, M.A., Hosseinkhani, M., 2011. Toxicity
evaluations of superparamagnetic iron oxide nanoparticles: cell “vision”
versus physicochemical properties of nanoparticles. ACS Nano 5, 7263–7276.
Mahmoudi, M., Lohse, S.E., Murphy, C.J., Fathizadeh, A., Montazeri, A., Suslick, K.S.,
2013b. Variation of protein corona composition of gold nanoparticles
following plasmonic heating. Nano Lett. 14, 6–12.
Mahmoudi, M., Saeedi-Eslami, S.N., Shokrgozar, M.A., Azadmanesh, K., Hassanlou,
M., Kalhor, H.R., et al., 2012. Cell “vision”: complementary factor of protein
corona in nanotoxicology. Nanoscale 4, 5461–5468.
Mahon, E., Salvati, A., Baldelli Bombelli, F., Lynch, I., Dawson, K.A., 2012. Designing
the nanoparticle–biomolecule interface for “targeting and therapeutic
delivery”. J. Control. Rel. 161, 164–174.
Marchini, C., Pozzi, D., Montani, M., Alfonsi, C., Amici, A., Candeloro De Sanctis, S.,
et al., 2011. Role of temperature-independent lipoplex-cell membrane
interactions in the efficiency boost of multicomponent lipoplexes. Cancer Gene
Ther. 18, 543–552.
Monopoli, M.P., Bombelli, F.B., Dawson, K.A., 2011a. Nanobiotechnology:
nanoparticle coronas take shape. Nat. Nanotechnol. 6, 11–12.
Monopoli, M.P., Walczyk, D., Campbell, A., Elia, G., Lynch, I., Baldelli Bombelli, F.,
et al., 2011b. Physical–chemical aspects of protein corona: relevance to in vitro
and in vivo biological impacts of nanoparticles. J. Am. Chem. Soc. 133,
2525–2534.
Prapainop, K., Witter, D.P., Wentworth Jr., P., 2012. A chemical approach for
cell-specific targeting of nanomaterials: small-molecule-initiated misfolding
of nanoparticle corona proteins. J. Am. Chem. Soc. 134, 4100–4103.
Röcker, C., Pötzl, M., Zhang, F., Parak, W.J., Nienhaus, G.U., 2009. A quantitative
fluorescence study of protein monolayer formation on colloidal nanoparticles.
Nat. Nanotechnol. 4, 577–580.
Sallam, R.M., 2015. Proteomics in cancer biomarkers discovery: challenges and
applications. Dis. Markers.
Salvati, A., Pitek, A.S., Monopoli, M.P., Prapainop, K., Bombelli, F.B., Hristov, D.R.,
et al., 2013. Transferrin-functionalized nanoparticles lose their targeting
capabilities when a biomolecule corona adsorbs on the surface. Nat.
Nanotechnol. 8, 137–143.
Tenzer, S., Docter, D., Kuharev, J., Musyanovych, A., Fetz, V., Hecht, R., et al., 2013.
Rapid formation of plasma protein corona critically affects nanoparticle
pathophysiology. Nat. Nanotechnol. 8, 772–781.
Walkey, C.D., Chan, W.C.W., 2012. Understanding and controlling the interaction of
nanomaterials with proteins in a physiological environment. Chem. Soc. Rev.
41, 2780–2799.
Walkey, C.D., Olsen, J.B., Guo, H., Emili, A., Chan, W.C., 2012. Nanoparticle size and
surface chemistry determine serum protein adsorption and macrophage
uptake. J. Am. Chem. Soc. 134, 2139–2147.
Zhang, H., Burnum, K.E., Luna, M.L., Petritis, B.O., Kim, J.S., Qian, W.J., et al., 2011.
Quantitative proteomics analysis of adsorbed plasma proteins classifies
nanoparticles with different surface properties and size. Proteomics 11,
4569–4577.
Zheng, T., Pierre-Pierre, N., Yan, X., Huo, Q., Almodovar, A.J., Valerio, F., et al., 2015.
Gold nanoparticle-enabled blood test for early stage cancer detection and risk
assessment. ACS Appl. Mater. Interfaces 7, 6819–6827.