BASIC SCIENCE

Nanomedicine: Nanotechnology, Biology, and Medicine

14 (2018) 195 – 204

Original Article nanomedjournal.com

Engineering macrophage-derived exosomes for targeted paclitaxel

delivery to pulmonary metastases: in vitro and in vivo evaluations
Myung Soo Kim, PhD a, b , Matthew J. Haney a, b , Yuling Zhao a, b , Dongfen Yuan a, b ,
Irina Deygen, PhD c , Natalia L. Klyachko, PhD a, b, c , Alexander V. Kabanov, PhD a, b, c ,
Elena V. Batrakova, PhD a, b,⁎
a
Center for Nanotechnology in Drug Delivery, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
b
Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC, USA
c
Deparment of Chemical Enzymology, Faculty of Chemistry, M.V. Lomonosov Moscow State University, Moscow, Russia
Received 20 April 2017; accepted 19 September 2017

Abstract

Exosomes have recently emerged as a promising drug delivery system with low immunogenicity, high biocompatibility, and high efficacy
of delivery. We demonstrated earlier that macrophage-derived exosomes (exo) loaded with a potent anticancer agent paclitaxel (PTX)
represent a novel nanoformulation (exoPTX) that shows high anticancer efficacy in a mouse model of pulmonary metastases. We now report
the manufacture of targeted exosome-based formulations with superior structure and therapeutic indices for systemic administration. Herein,
we developed and optimized a formulation of PTX-loaded exosomes with incorporated aminoethylanisamide-polyethylene glycol (AA-PEG)
vector moiety to target the sigma receptor, which is overexpressed by lung cancer cells. The AA-PEG-vectorized exosomes loaded with PTX
(AA-PEG-exoPTX) possessed a high loading capacity, profound ability to accumulate in cancer cells upon systemic administration, and
improved therapeutic outcomes. The combination of targeting ability with the biocompatibility of exosome-based drug formulations offers a
powerful and novel delivery platform for anticancer therapy.
Published by Elsevier Inc.

Key words: Drug delivery systems; Exosomes; Paclitaxel; Pulmonary metastases

The lung is one of the most common sites of metastases and
tumor relapse following treatment of the primary tumor mass;
lung metastases are identified in 30%-55% of all cancer patients.
Because pulmonary metastases are distributed throughout the
pulmonary parenchyma, their excision is difficult, if not
impossible. Clinical reports have highlighted a rare occurrence
of complete regression for metastatic tumors in patients who
underwent standard-of-care palliative treatment involving radi-
ation or immunotherapy. As a result, lung cancer is the leading
cause of cancer-related deaths in the world, responsible for more
deaths than breast, colon, and prostate cancers combined. 1 In
particular, non-small cell lung cancer (NSCLC) is responsible for
~85% of all lung cancers, and prognosis of metastatic NSCLC is
poor; chemotherapy provides only minimal increases in survival
rates, with a five-year survival rate of less than 15%. 2,3 Thus, the
efficient, targeted delivery of anticancer agents to pulmonary
metastases remains one of the greatest challenges for the therapy.
The recently emerged field of nanotechnology holds great
promise for developing drug delivery systems with targeting and
controlled-release characteristics for cancer treatment; there have
been many new advances and innovations made in this field
during the past decade. 1 A large proportion of chemotherapeutic
drugs have low aqueous solubility, consequently requiring the
use of specialized delivery vehicles (e.g. micelles, liposomes,
polymeric nanoparticles, or other types of nanoparticles) for
parenteral administration. However, these nanosized delivery

This study was supported by the United States National Institutes of Health grants 1RO1 NS057748 and The Carolina Partnership, a strategic partnership
between the UNC Eshelman School of Pharmacy and The University Cancer Research Fund through the Lineberger Comprehensive Cancer Center (to EVB),
RR021937 (to AVK), and grant RSF-14-13-00731 (to both AVK and NLK).
⁎Corresponding author at: UNC Eshelman School of Pharmacy, University of North Carolina at Chapel Hill, Chapel Hill, NC 27599-7362.
E-mail address: batrakov@ad.unc.edu (E.V. Batrakova).

https://doi.org/10.1016/j.nano.2017.09.011
1549-9634/Published by Elsevier Inc.

Please cite this article as: Kim MS, et al, Engineering macrophage-derived exosomes for targeted paclitaxel delivery to pulmonary metastases: in vitro and
in vivo evaluations. Nanomedicine: NBM 2018;14:195-204, https://doi.org/10.1016/j.nano.2017.09.011
196 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204
vehicles are often difficult to manufacture and may cause
unwanted side effects (such as the excipient Cremophor EL in
the commercial formulation of PTX, Taxol). In addition,
conventional drug nanoformulations normally are cleared
rapidly from the circulation by the mononuclear phagocyte
system (MPS). 4 This demands innovative approaches for
engineering drug delivery systems.
Exosomes are membrane-derived vesicles ~40-200 nm in
diameter 5; they may be found in extracellular bodily fluids and in
conditioned cell culture media. 6 Exosomes are released by a
variety of cell types; and formed when multi-vesicular bodies
inside the cells fuse with the plasma membrane and release
intraluminal vesicles into the extracellular microenvironment. 6,7
Exosomes naturally function as intracellular messengers,
carrying RNAs and proteins between the cells. 8 Recently,
exosomes have begun to be explored for use as drug delivery
vehicles for non-native therapeutics such as nucleic acids, 9–14
therapeutic proteins, 15 and small molecule drugs. 6,16,17 The
membranotropic nature of exosomes suggests that these drug
carriers will be able efficiently interact with the target cancer
cells and deliver their toxic payload. This process is facilitated by
membrane interactions and fusion due to the expression of
various adhesive proteins (tetraspanins and integrins) on the
surface of exosomes. 18 Noteworthy, it was reported that
exosomes are able to deliver their intraluminal cargo into the
cytosol of target cells. 19 In addition, allogenic exosomes have an
immune privileged status, which allows for decreased drug
clearance by MPS. Specifically, immunocytes-derived exosomes
are known to express CD47 receptor, 20 which interacts with
signal regulatory protein α (SIRPα) to produce a “don't eat me”
signal in phagocytes. 21 These unique features make exosomes
an attractive option for use as a drug delivery vehicle for
cancer treatment.
We reported earlier that macrophage-derived exosomes can
be loaded with low molecular chemotherapeutics, such as PTX,
and doxorubicin (DOX), 22 or therapeutic proteins, such as
catalase 15 or brain derived neurotrophic factor, BDNF. 23
Different loading procedures (co-incubation with a drug at
room temperature, electroporation, saponin permeabilization,
extrusion, freeze–thaw cycles, or sonication) were utilized to
incorporate these therapeutic agents into exosomes. We
demonstrated that a reformation of exosomal membranes upon
sonication allowed efficient drug loading along with preservation
of exosomal structure and internal content. Here we report the
development of improved exosome-based PTX formulation
targeted to cancer cells for systemic administration. It has been
shown that a variety of cancer types, including NSCLC,
overexpress sigma receptor, 24 a membrane-bound protein with
an as-yet undefined role. AA is a ligand with high affinity for
sigma receptor and has been utilized to targeted delivery of Dox,
proteins, and siRNA. 25–27 Next, the most common method of
reducing the immunogenicity of nanoformulated drugs is to
decorate the nanoparticle in a polyethylene glycol (PEG) corona,
which reduces recognition by the MPS and aids in avoiding
clearance. To this end, Kooijmans et al have recently shown that
the introduction of polyethylene glycol to exosomes results in
stealth properties, which significantly increases their circulation
time in mice. 28
Based on these findings, we hypothesized that modification
of PTX-loaded exosomes with PEG-AA vector moiety will
improve their circulation time in the blood and allow to target
pulmonary metastases. Our investigations revealed a robust
accumulation and nearly complete co-localization of systemi-
cally administered AA-exosomes with pulmonary metastases,
and greater in vivo therapeutic efficacy as compared to
non-vectorized exoPTX and Taxol. Thus, exosome-based
formulations may represent the next generation of drug delivery
systems that combines nanoparticle size with targeted drug
delivery and low immunogenic profile.
Materials and methods
Description of reagents, cell culture conditions, animals used
in these studies, characterization of exosomal formulations as
well as accumulation and intracellular trafficking experiments
and statistical analysis are described in Supplementary Material.
Preparation of AA-vectorized exosomes targeted to the sigma
receptor
For all in vitro experiments, exosomes were harvested from
the supernatants of RAW 264.7 cells cultured in
exosome-depleted media using the ExoQuick-TC™ Kit as
described earlier. 22 For all in vivo experiments, exosomes
released by primary bone-marrow derived macrophages (BMM)
isolated from C57BL/6 mice were utilized. Exosomes targeted to
sigma receptor with DSPE-PEG-AA (AA-PEG-exo) and
non-vectorized control exosomes with DSPE-PEG (PEG-exo)
were prepared as follows: exosomes were isolated from
macrophage media as described above and then DSPE-PEG or
DSPE-PEG-AA (50 μg/ml) was added to the mixture (for
PEG-exo and AA-PEGexo, respectively). 100 μL PTX (10 mg/mL
in ethanol) was also added to the exosomes to incorporate PTX
using sonication method that was developed in our laboratory. 22
After sonication, AA-PEGexo, or PEGexo, or AA-exoPTX
solutions were incubated at 37 °C for 60 min to allow for recovery
of the exosomal membrane. Exosomes were purified from the
excess of free DSPE-PEG or DSPE-PEG-AA by size exclusion
chromatography using a NAP-10 Sephadex G25 column (GE
Healthcare, Buckinghamshire, UK). Purified exosomal formula-
tions were collected and stored at −20 °C.
Drug loading and optimization AA-PEG-exoPTX
For PTX loading into vectorized exosomes, 1 mL of purified
exosomes (~10 11 exosomes) in PBS was mixed with PTX and
DSPE-PEG-AA. For this purpose, first PTX (10 mg/mL drug in
EtOH stock solution) was added to 1 mL exosomes in PBS.
Then, different amounts of AA-PEG-DSPE (0.05-0.50 mg/ml) in
PBS were added to the mixture of exosome with PTX. The
obtained mixture was sonicated using a Model 505 Sonic
Dismembrator using .25″ tip (Thermo Fisher Scientific, USA) as
described earlier. 22 After the sonication, a solution
AA-vectorized exosomes loaded with PTX (AA-PEG-exoPTX)
were incubated at 37 °C for 60 min to allow for recovery of the
exosomal membrane. The excess free PTX and AA-PEG-DSPE
 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204
were eliminated from AA-PEG-exoPTX by size exclusion
chromatography as described elsewhere. 22 The amount of PTX
loaded into exosomes was measured by HPLC method. 22 All
analyses were performed using a C18 column (Supelco Nucleosil
C18, 250 mm × 4.6 mm, 5 μm, 100 Å, Sigma-Aldrich,) with a
mobile phase of H2O:acetonitrile (45:55, v/v) at a flow rate of
1 mL/min at 30 °C. Loading capacity is expressed by μg protein
of exosomes.
Biodistribution of intravenously injected exosomes in mice with
pulmonary metastases
To utilize fluorescence imaging, 3LL-M27 cells were trans-
duced with lentiviral vectors encoding the optical reporter FITC
(FITCFlmC, green) fluorescent protein as reported earlier. 29 The
viral construct also encoded for a puromycin resistance gene
downstream of FITC, which was introduced to enable for the
selection of positively transduced cells. C57BL/6 mice were
injected intra tail vein (i.v.) with FITC-FLmC-3LL-M27 cells
(5 × 10 6 cells/mouse in 100 μl saline) and pulmonary metastases
were allowed to establish. In parallel, exosomes isolated from
autologous macrophages conditioned media were stained with DiL
(red) and vectorized to sigma receptor with DSPE-PEG-AA. To
stain exosomes with DiL, exosome pellet was rehydrated in 500 μL
PBS, supplemented with 1 mM stock solution DiL in DMSO
(5 μL), and incubated for 20 minutes at RT. Following incubation,
exosomes were isolated by PEG precipitation and rehydrated in
500 μL PBS. The fluorescently-labeled exosomes were addition-
ally purified from remaining non-incorporated DiL on NAP 10
column. Twelve days following cancer cells injection, mice with
established lung metastases were intravenously (i.v.) injected with
autologous vectorized exosomes DiL-AA-PEG-exo, or
non-vectorized exosomes DiL-exo as a control group (10 8
particles/100 μl, n = 4 per group). Four hours later, mice were
sacrificed, perfused, lungs were extracted and sectioned at a
thickness of 20 μm; nuclei were stained with DAPI (300 mM, 5
min). The images of lung and liver sections were examined by a
confocal fluorescence microscopic system ACAS-570 (Meridian
Instruments, Okimos, MI) with argon ion laser and corresponding
filter set. Digital images were obtained using the CCD camera
(Photometrics) and Adobe Photoshop software. Quantification of
immunostaining was performed with ImageJ software, utilizing
JACoP plugins to calculate Pearson's co-localization
coefficients. 30 The comparison was performed in randomized
order of the blinded experiment on 10-15 sets of images acquired
with the same optical settings.
Therapeutic efficacy of AA-PEG-exoPTX against pulmonary
metastases
The antineoplastic effects of PTX exosome formulation were
evaluated in a mouse model of pulmonary metastases with
3LL-M27 cells transduced with lentiviral vectors encoding the
optical reporter mCherry (GBM8FlmC, red) as described
earlier. 22 To establish pulmonary metastases, C57BL/6 mice
were i.v. injected with 8FlmC-3LL-M27 cancer cells (5 × 10 6
cells/100 μl/ mouse). Forty-eight hours later, mice were treated
i.v. with autologous AA-PEG-exoPTX, or exoPTX, or empty
exosomes (exo) (4 × 10 11 particles/100 μl, three times on day 1,
197
4, and 7; 0.5 mg/kg totally), or Taxol (same regiment as
exosome-based formulations), or saline as a control (n = 7). To
assess amount of cancer metastases, two mice from each group
were sacrificed on day 18, perfused, and lung slides were
examined by confocal microscopy. The rest of the mice (n = 5)
were monitored daily for the signs of the reduced physical
activity and the progression of the tumor. The survive time of
each mouse was recorded. To perform quantitative analysis, an
additional experiment with the same treatment groups of mice
(n = 5) was carried out. On day 18 mice were sacrificed,
perfused, and lung slides were examined by confocal microsco-
py. The metastases area on the slides was calculated using
ImageJ software. The comparison was performed in randomized
order of the blinded experiment on 10-15 sets of images acquired
with the same optical settings.
Statistical analysis
For the all experiments, data are presented as the mean ±
S.E.M. Tests for significant differences between the groups
were performed using a t-test or one-way ANOVA with multiple
comparisons (Fisher's pairwise comparisons) using GraphPad
Prism 5.0 (GraphPad software, San Diego, CA, USA). A
minimum P value of 0.05 was chosen as the significance level.
Results
Manufacture and characterization of AA-PEG-exoPTX
We demonstrated earlier that efficient loading of PTX can be
achieved, when exosomes are subjected to ultrasound treatment
in the presence of the drug. 22 Herein, we applied the same
approach for simultaneous loading of PTX, and vectorization of
exosomes to sigma receptor using AA moiety. Obtained
formulations were characterized by size, charge, protein content,
and PTX loading capacity (LC). First, the effect of incorporation
of vector lipid molecule (DSPE-PEG-AA) on the LC was
evaluated. LC was expressed as the amount of the drug vs. the
amount of exosomal protein. HPLC analysis revealed that
incorporation of high amounts of AA-PEG-DSPE (0.5 mg/ml) into
exosomes significantly decreased their LC for PTX (Figure 1, A). In
contrast, lower amounts of the lipid (0.25-0.05 mg/ml) did not
affect LC for PTX. It is likely that excess of hydrophobic chains
of the lipid incorporated into exosomal membranes diminished
available for PTX space. Therefore, we choose the high amount
of the lipid (0.25 mg/ml) that did not significantly reduce PTX
loading to vectorize exosomes. LC for the optimal
AA-PEG-exoPTX formulation was ~33%, comparable to the
LC achieved for non-vectorized exoPTX (Figure 1, A).
Noteworthy, ultrasound treatment significantly increased
amount of PTX incorporated into exosomes; the LC in
exosomes without sonication was as low as 1.4%. 22
To address a concern about possible alterations in the exosomal
membranes upon incorporation of vector moiety (AA-PEG-DSPE)
using sonication, the levels of exosome-specific proteins, TSG101
and flotillin, in different exosomal formulations were examined by
western blot (Figure 1, B). The mild sonication utilized for PTX
loading with six cycles, and intermediate time out for cooling down
198 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204

Figure 1. Characterization of exoPTX and AA-PEG-exoPTX formulations. PTX and vector moiety AA-PEG-DSPE were incorporated in exosomes by
sonication procedure. (A) Loading Capacity (LC) for PTX in vectorized exosomes was measured by HPLC. Increasing amounts of vector moiety resulted in LC
decrease. The highest concentration of AA-PEG-DSPE that did not lower the LC for PTX in exosomes was chosen for subsequent experiments (shown by
arrow). (B) Western blot data indicated that formulations retained the exosome markers TSG101, flotillin, and LFA1, specific marker for lymphocytes, after the
sonication procedure. (C) Size was measured by NTA and DLS; the zeta potential was measured by DLS. The data were obtained from five independent
experiments.

and membrane restoration did not affect the protein content of
exosomes. In particular, sonicated vectorized and non-vectorized
exosomes, as well as vectorized exosomes loaded with PTX
showed elevated expression of exosome-associated proteins
(TSG101, and flotillin) as compared to cell lysate, which displayed
greater levels of β-actin (Figure 1, B). Noteworthy, we reported
earlier that sonication itself did not affect levels of
exosome-specific proteins compared to naïve non-sonicated
exosomes. 22 Furthermore, exosomal formulations, as well as
parental macrophages, were also found to express the lymphocyte
function associated antigen-1 (LFA1, subunit CD11a) (Figure 1, B),
which assists in cell uptake and may bind to endothelial cell adhesion
molecules overexpressed on activated endothelial cells, such as those
found in tumors. 31 This is important, since the presence of LFA1 on
the surface may improve specific targeting of exosome-based PTX
formulations to cancer cells.
Finally, the hydrodynamic size was determined by DLS and
NTA (Figure 1, C). Naïve empty exosomes had a narrow size
distribution, with an average particle diameter of 110.8 ± 4.1 nm
and 75.9 ± 2.6 nm as revealed by NTA and DLS, respectively.
The sonication procedure significantly increased size of
exosomes up to 291 ± 3.5 nm (Figure 1, C). Noteworthy,
exosomes sonicated in the presence of PTX were smaller than
those sonicated without PTX. This effect may be due to the
stabilization of exosomal membranes by the incorporated drug.
Next, it is known that sialic acid on glycosylated lipids and
proteins is abundant on cell membranes and contributes to the
surface charge of individual cellular membranes. In this regard,
loading of exosomes with PTX did not significantly alter the
slightly negative change of the nanocarriers (Figure 1, C).
However, vectorized AA-PEG-exoPTX formulation was found
to have a less negative surface charge than exoPTX, probably
due to the shielding of the exosomal membrane by the long PEG
chains of the lipid (Figure 1, C).
Effect of AA-PEG-DSPE incorporation on membrane fluidity in
exosomes
Fluidity of exosomal membrane upon AA-PEG-DSPE
incorporation was examined using BODIPY-PC. This is a
hydrophobic fluorescent compound, which incorporates in the
hydrocarbon regions of lipid membranes. Transfer of
BODIPY-PC from the aqueous environment into lipid bilayers
results in a drastic increase of the fluorescence emission for this
probe. Once the probe is incorporated into lipid membranes, its
fluorescence polarization (reflecting the ability to freely rotate in
the lipid biolayers) strongly depends on the microenvironment,
with decreases in membrane microviscosity resulting in
increased fluorescence polarization. In this study,
BODIPY-PC-labeled exosomes were sonicated in the presence
of AA-PEG-DSPE lipid, and fluorescence polarization was
recorded (Figure 2). Co-incubation of PTX with exosomes in the
absence of sonication did not alter membrane microviscosity;
however, small but statistically significant fluidization of
exosomal membranes was recorded when a high amount of
lipid (0.5 μg/ml) was added to the solution. Next, significant
decreases (more than two times) in membrane microviscosity
were recorded upon sonication that is consistent with our
previous observations. 15 The fluidity of exosomal membranes
was partially restored when PTX was added to the solution.
Sonication of exoPTX in the presence of the lipid further
increased membrane microviscosity up to naïve non-sonicated
exosomes (Figure 2). Noteworthy, the greater amount of lipid
was added to the solution; the higher the microviscosity levels
obtained. We hypothesize that the sonication leads to
 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204
Figure 2. Effect of AA-PEG-DSPE incorporation and PTX loading on
fluidity of exosomal membranes. BODIPY-PC-labeled exosomes were
examined by fluorescence polarization measurements. The ultrasound
treatment significantly decreased microviscosity of exosomal membranes
compared to naïve exosomes. The microviscosity of sonicated exosomes was
increased with PTX loading and then further increased with incorporation of
the lipid upon sonication. Values are means ± SEM (n = 4). Symbols indicate
the relative level of significance compared with naïve exosomes (P b 0.05).
The data were obtained from three independent experiments. Symbols
indicate the relative level of significance compared to protein-striped
exosomes (*P b 0.05, **P b 0.005).
dysregulation of exosomal membranes and creation of additional
space for PTX molecules. This resulted in an increased LC for
PTX. The incorporation of high amounts of lipid molecules upon
sonication allowed sealing membrane bilayers that may impede
PTX loading, and as a result, diminish LC (Figure 1, A).
Accumulation of AA-PEG-exoPTX in target cancer cells in vitro
The ability to deliver the drug payload into target cells is
crucial for the therapeutic efficacy of exosomal formulations.
Although the molecular function of sigma receptors is not yet
fully defined, there is increasing evidence that these receptors are
overexpressed in many cancer cells. 24 First, we validated sigma
receptor as a target for LLC murine model of pulmonary
metastases. Western blot analysis revealed high expression levels
in murine LLC cells (3LL-M27), as well as murine lung
adenocarcinoma cells (344SQ), human small-cell lung carcino-
ma cells (H69/AR), and human non-small cell lung carcinoma
cells (A549), and low, if any, in normal human lung fibroblasts
(Hel 299) (Supplemental Figure 1, A). The protein bands were
quantified and normalized to beta actin (Supplemental Figure. 1, B).
This indicates that the choice of AA (a sigma receptor ligand) as a
vector to target 3LL-M27 cells is appropriate for the presented
experiments in the murine LLC model, as well as for further clinical
evaluations.
Previously, we demonstrated that accumulation levels of
fluorescently-labeled exosomes in 3LL-M27 cells was consid-
erably greater (about 30 times) then accumulation of liposomes
or polystyrene nanoparticles. 15 Herein, the receptor-mediated
199
accumulation of DiL-labeled vectorized exosomes (AA-PEG-exo)
was studied in target 3LL-M27 cells in vitro, and compared against
control non-vectorized sonicated exosomes (exo), as well as
exosomes with incorporated PEG-DSPE lipid without AA
targeting moiety (PEG-exo) (Figure 3, A). The obtained data
indicated that vectorized AA-PEG-exo nanocarriers were taken up
in much higher quantities than non-vectorized sonicated exo-
somes. The PEGylated exosomes without AA-targeting moiety
were taken up less than parental exosomes, probably due to the
PEG chains blocking interaction of exosomal surface proteins,
which assist in cell accumulation.
To further assess the capability of incorporated AA to target
exosomes to sigma receptor, a competitive inhibition study was
carried out in 3LL-M27 cells (Figure 3, B). In this experiment,
3LL-M27 cells were pre-treated with free AA at varying
concentrations, washed with PBS, and then equal amounts of
vectorized AA-PEG-exo along with free AA were added to the
cells for one hour. Fluorescence levels were measured; the
amount of exosomes/μg protein was quantified and graphed
against the concentration of AA (Figure 3, B). Results showed a
dose-dependent response of AA-PEG-exo to competitive
inhibition by increasing concentrations of free AA, indicating
that AA-PEG-exo were taken up by receptor-mediated endocy-
tosis. Noteworthy, even a large amount of free AA added to the
AA-vectorized exosomes was not able completely inhibit
exosome uptake in target cells, suggesting involvement of
other exosomal surface proteins in this process, for example
LFA1 (as demonstrated by western blot, Figure 1, B) that is
consistent with our recent report. 23
The importance of exosomal surface proteins in assisting in
exosome take-up was confirmed in accumulation studies using
proteinase K treatment to strip exosomes of surface proteins
(Figure 4). In this experiment, DiL-labeled sonicated
AA-vectorized (AA-PEG-exo) and non-vectorized sonicated
exosomes (exo/sonic), as well as non-sonicated naïve exosomes
(exo/naive) were incubated with 3LL-M27 cancer cells for
various times; and the number of taken exosomes was accessed
by fluorescence. The obtained data indicate that accumulation
levels in target cells increased in order: exo/naïve b exo/sonic b
AA-PEG-exo (Figure 4). This confirmed our previous reports
that treatment with ultrasound improved exosome accumulation
in cancer cells, 22 as well as neuronal PC12 cells. 15 In parallel,
the same exosomal formulations were treated with proteinase K,
and added to the cells. The digestion of the exosomal surface
proteins significantly decreased uptake by target cells in all
formulations. These results clearly show the advantages of
exosome-based drug delivery systems over common synthetic
nanocarriers related to the facilitated uptake of exosome carriers
by means of surface adhesive proteins. Noteworthy, stripping of
surface proteins from vectorized exosomes (AA-PEG-exo)
decreased their transport at significantly lesser extent than
non-vectorized exosomes (Figure 4), probably due to the assisted
AA-mediated accumulation in cancer cells.
Finally, to assess the effect of AA-vectorization on
intracellular trafficking of exosomes, 3LL-M27 cells were
supplemented with DiL-labeled AA-PEG-exo and exo as a
control, and then stained with MitoTracker, or LysoTracker, or
ERTracker dyes to visualize mitochondria, lysosomes, or
200 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204

Figure 3. Accumulation of AA-vectorized exosomes in cancer cells. (A) 3LL-M27 cells were incubated with fluorescently-labeled AA-PEG-exo, or PEG-exo, or
exo for various times, and accumulation levels were measured. AA-PEG-exo were more readily taken up by 3LL-M27 as compared to exo and PEG-exo. (B)
AA-PEG-exo formulation showed a dose-dependent response to competitive inhibition by AA, indicating that this formulation entered cells by receptor
mediated endocytosis. Results are expressed as number of exosomes/μg protein vs. concentration of AA. The data were obtained from three independent
experiments.

investigated the percentage of exosomes co-localized with

Figure 4. Importance of surface proteins on exosome accumulation in
3LL-M27 cancer cells. Naïve (circle), sonicated (triangle), and AA-vector-
ized (cross) exosomes were incubated with Proteinase K to strip surface
proteins and subsequently labeled with DiL; accumulation of Proteinase
K-treated exosomes (dashed line) or control exosomes (solid line) was
examined in cancer cells. Stripping the exosomal surface proteins resulted in
significant decreases in exosome accumulation levels in target cells for all
formulations. The data were obtained from three independent experiments.
endoplasmic reticulum (ER), respectively (Supplemental Figure 2).
Confocal images revealed that exosomes preferentially distribute in
order: lysosomes N ER N mitochondria. Noteworthy, the intracel-
lular fate of exosomes was not altered by the addition of a vector to
the sigma receptor.
Co-localization of systemically-administered exosomes with
pulmonary metastases in LLC mice
To assess the ability of AA-vectorized exosomes to target
pulmonary metastases, studies were conducted in the LLC
mouse model with 3LL-M27 cells overexpressing the optical
reporter green fluorescent protein (GFP). In particular, we
cancer cells in the lungs. To induce metastases, C57BL/6 mice
were injected with GFP/3LL-M27 (5 × 10 6 cells/100 μL) as
described above. Twenty-one days later, DiL-labeled autologous
non-vectorized (exo) and AA-vectorized (AA-PEG-exo) exo-
somes were injected i.v. to C57BL/6 tumor-bearing mice. Four
hours later, mice were sacrificed, perfused; lungs were sectioned
on microtome and examined by confocal microscopy. Confocal
images revealed 94.4 ± 0.8% of AA-exosomes in immunohis-
tochemistry slides were co-localized with lung metastases
(Figure 5, D-F) indicating efficient targeting of AA-exoPTX
in vivo. In contrast, only 21.8 ± 0.2% of systemically-injected
non-vectorized exosomes were co-localized with cancer metas-
tases (Figure 5, A-C). Noteworthy, no AA-exosomes were
found in the lungs of healthy animals (Figure 5, G-J). This
suggests that systemically-administered AA-exosomes can
efficiently reach pulmonary metastases and deliver their drug
payload to target cancer cells.
It was reported that along with pulmonary metastases, i.v.
injected 3LL- M27 cells populate liver in C57BL/6 mice. 32 We
investigated, whether AA-vectorized exosomes were able to
target liver metastases similar to pulmonary metastases. As
expected, confocal images showed close to complete
co-localization of fluorescently-labeled AA-PEG-exosomes
with cancer cells in the liver (Supplemental Figure 3) suggesting
that AA-exosomal formulations may be suitable for other than
pulmonary metastases therapy.
Therapeutic efficacy of AA-PEG-exoPTX against lung metastases
C57BL/6 mice with established mCherry-3LL-M27 (red)
metastases were systemically injected with AA-PEG-exoPTX.
Mice injected with exoPTX, or Taxol, or saline, or exosomes
alone (without PTX) were used in control groups. Eighteen days
later mice were sacrificed, perfused, and lung slides were
examined by confocal microscopy (Figure 6, A). The images
demonstrated a superior antineoplastic efficacy of AA-exoPTX
compared to non-vectorized exoPTX, or Taxol that resulted in
 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204 201

Figure 5. Co-localization of systemically-delivered AA-exosomes with pulmonary metastases. Exosomes were isolated from macrophages conditioned media
and labeled with DiL dye (red, A and D). C57BL/6 mice were i.v. injected with 3LL-M27 cells transduced with lentiviral vectors encoding the optical reporter
GFP fluorescent protein (green, B and E). 7 days later, tumor-bearing mice were i.v. injected with DiL-labeled non-vectorized exosomes (red, A), or
AA-exosomes (red, D). Four hours later, mice were euthanized, perfused, lungs were sectioned, and stained with DAPI (blue). Confocal images revealed
significant co-localization of AA-exosomes with metastases (94.4 + 0.8%, F) that was greater than those of non-vectorized exosomes (21.8 + 0.2%, C). No
exosomes were found in lungs of healthy animals without metastases (G-J). Bar: 20 μm. The data were obtained from three independent experiments.

the potent eradication of pulmonary metastases. Additional Discussion

images of randomly selected lung slides are shown on
Supplemental Figure 4. The quantitative assessment of metas-
tases area on the lung slides of animals is shown on Figure 6, B.
The survival studies confirmed these results (Figure 6, C).
Administration of AA-exoPTX formulation caused a signifi-
cantly stronger suppression of metastases growth and greater
survival time. Noteworthy, injections of control empty (naïve)
exosomes without PTX incorporated did not slow down tumor
growth in mice with pulmonary metastases. This confirms the
superior antineoplastic efficacy of AA-PEG-exoPTX upon
systemic administration, as compared to Taxol, or
non-vectorized exoPTX formulation.
The efficient targeted delivery of anticancer agents to
pulmonary metastases remains one of the greatest challenges
for the treatment of lung cancer. Several nanoformulations are
being studied in clinical trials, or have already been approved by
the FDA for use in humans. 33,34 However, conventional
nanoparticles have limited biocompatibility, and normally are
cleared rapidly from the circulation by the MPS. 4 We have
developed a new drug delivery system that is based on natural
vectors, exosomes, released by autologous macrophages for the
delivery of PTX. We utilize macrophage-derived exosomes that
exert unique biological activity reflective of their origin, and
202 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204

Figure 6. AA-PEG-ExoPTX induced potent anticancer effect against lung metastases in LLC mouse model. (A) C57BL/6 mice with established Luc/
mCherry-3LL-M27 lung metastases were i.v. treated with: saline, or Taxol, or exoPTX, or AA-PEG-exoPTX, or exosomes alone (no drug). 18 days later, mice
were sacrificed, perfused, and lungs slides were examined by confocal microscopy. AA-PEG-exoPTX treatment resulted in a potent inhibition of metastases
(red) that was more effective than treatment with Taxol or exoPTX. The bar: 50 μm. (B) Quantitative assessment of metastases area on the lung slides of animals
treated with: 1) saline, or 2) Taxol, or 3) exoPTX, or 4) AA-PEG-exoPTX, or 5) exosomes alone. **P b 0.005. (C) A survival of C57BL/6 mice with established
metastases was recorded for four treatment groups: 1) saline (dimonds), or Taxol (squares), or 3) exoPTX (triangles), or 4) AA-PEG-exoPTX (crosses), or 5)
exosomes alone (stars). A superior effect on suppression of metastases growth and greater mice survival was recorded in AA-PEG-exoPTX treatment group
(n = 6). The data were obtained from three independent experiments.

therefore, can provide advantages of both nanotechnology, and
cell-mediated drug delivery that is based on the innate functions
of immune cells.
Using exosomes as drug delivery vehicles offers a number of
benefits over common drug administration regimens; however,
there are some limitations and challenges that need to be
addressed. One of the major challenges is the efficient loading of
exosomes without significant alterations to the structure and
content of exosomal membranes. PTX is a highly hydrophobic
compound that is likely to be incorporated into the inner region
of the relatively tight and highly structured lipid bilayers of
exosomes. Therefore, we developed a specific procedure
wherein lipid bilayers are reshuffled upon mild sonication. To
target exosomal carriers to cancer cells, we incorporated a vector
moiety with AA that is known to specifically bind to the sigma
receptor using the same sonication procedure. Herein, we
optimized the amount of incorporated AA vector moiety that
did not impede with the loading of PTX. The obtained
AA-PEG-exoPTX formulation showed an extraordinary ability
to accumulate in target cancer cells; these exosomes were taken
up via receptor-mediated endocytosis in considerably greater
numbers than non-vectorized exosomes in vitro. Noteworthy,
addition of AA moiety did not alter intracellular trafficking of
exosomal formulations in cancer cells.
The most interesting results were obtained in the mouse LLC
model. Our data demonstrate a robust accumulation and nearly
complete co-localization of systemically administered
AA-vectorized exosomes with cancer metastases. We hypothe-
sized that both AA-vector, and LFA1 protein expressed on
exosomal membranes were responsible for this preferential
accumulation in pulmonary metastases upon systemic adminis-
tration. Significantly, exosomes were targeted to liver metastases
as well; a complete co-localization of AA-vectorized exosomes
with cancer cells was demonstrated in this work. Taking into
account that hydrophobic fluorescent dye DiL may serve as a
model drug that has a low solubility in water (for example PTX),
it suggests that exosomes can target cancer metastases and
deliver their cytotoxic payload specifically to these cells
avoiding healthy tissues. The investigations of trafficking of
exosomes and their components (loaded therapeutic agents, as
well as exosomal lipids, and proteins, etc.) are on the way in our
laboratory.
Finally, a systemic administration of AA-vectorized exoso-
mal formulation of PTX resulted in the superior antineoplastic
effect and increased survival times in mice with pulmonary
metastases, as compared to exoPTX or Taxol treated animals.
We hypothesized that this effect was due to the superior
therapeutic efficacy of the formulation. Specifically, the efficient
 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204
targeting of PTX incorporated into AA-vectorized exosomes
resulted in the potent inhibition of pulmonary metastases growth.
The confocal images and quantification of metastases area on
lung slides confirmed this suggestion. Furthermore, targeting of
PTX to the cancer cells may also decrease side effects of the
formulation that also will manifest in increased survival times.
Notable, LLC cells are known to express the Pgp drug efflux
transporter in vivo. 35 We demonstrated earlier that the
incorporation of PTX into exosomes may not only increase its
solubility, but also allow for overcoming of Pgp-mediated drug
efflux in resistant cancer cells. This effect may be attributed to
differences in internalization routes for exoPTX and Taxol.
Exosomes and micelles, such as those found in Taxol, enter
resistant cancer cells by endocytosis, however exosomes have a
superior accumulation due to the presence of adhesion proteins,
tetraspanins, integrins, immunoglobulins, proteoglycans, and
lectins, 36 which are not found on artificial nanocarriers.
Furthermore, exosomes consist of cellular membranes that may
fuse with the plasma and/or endocytic membranes and deliver
their cargo, bypassing Pgp-mediated efflux.
Moreover, it is known that exosome-mediated cell-to-cell
communication is key in the battle between cancer and the
immune system. 37 Thus, Parolini et al 38 showed that exosome
fusion with target cells occurs more efficiently under acidic
conditions, implying that exosomes may be taken up preferen-
tially by tumors (which have an acidic microenvironment) rather
than the surrounding healthy tissue. Finally, decoration of
exosomes with PEG chains may considerably increase their
circulation in the blood as was demonstrated earlier. 28
Regarding the elimination fate and blood circulation time, the
source of exosomes is of great importance. Thus, when
exosomes released by cancer cells, they efficiently accumulated
in MPS and quickly eliminated from the blood stream. 38,39
Contrary, exosomes released by primary fibroblast-like mesen-
chymal cells can bypass immune clearance by monocytes and
macrophages that results in their prolonged blood circulation.
The reported mechanism is related to the CD47 molecule that is
expressed on the surface of exosomes and its binding partners
thrombospondin-1 (TSP1) and signal regulatory protein α
(SIRPα) that produce the “don't eat me” effect in
macrophages. 40 This may also enhance specificity of exosomal
formulations to cancer cells in MPS-reach organs. Further
studies of the mechanism of these superior antineoplastic effects
are ongoing in our lab.
Overall, all four mechanisms mentioned here are likely to
have significant impact on AA-PEG-exoPTX anticancer activity,
i.e.: (i) vector-mediated preferential accumulation in cancer cells,
(ii) efficient delivery of incorporated cargo into target cancer
cells, (iii) bypassing Pgp-mediated drug efflux in resistant cancer
cells, and (iv) prolonged circulation time in the blood via
PEGylation. In conclusion, exosomes promise an unparalleled
efficacy in the treatment of many life-threatening conditions,
including those lacking effective pharmacotherapy.
Acknowledgments
We are grateful to Dr. Leaf Huang for the invaluable
comments and suggestions.
203
Appendix A. Supplementary data
Supplementary data to this article can be found online at
https://doi.org/10.1016/j.nano.2017.09.011.
References
1. Jemal A, Siegel R, Ward E, Murray T, Xu J, Thun MJ. Cancer statistics,
2007. Clin 2007;57(1):43-66.
2. Goffin J, Lacchetti C, Ellis PM, Ung YC, Evans WK. Lung Cancer
Disease Site Group of Cancer Care Ontario's Program in Evidence-
Based C. First-line systemic chemotherapy in the treatment of advanced
non-small cell lung cancer: a systematic review. J Thorac Oncol
2010;5(2):260-74.
3. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H,
et al. Gefitinib or chemotherapy for non-small-cell lung cancer with
mutated EGFR. Med 2010;362(25):2380-8.
4. Peng Q, Zhang S, Yang Q, Zhang T, Wei XQ, Jiang L, et al. Preformed
albumin corona, a protective coating for nanoparticles based drug
delivery system. Biomaterials 2013;34(33):8521-30.
5. Ratajczak J, Wysoczynski M, Hayek F, Janowska-Wieczorek A,
Ratajczak MZ. Membrane-derived microvesicles: important and under-
appreciated mediators of cell-to-cell communication. Leukemia
2006;20(9):1487-95.
6. Kalani A, Kamat PK, Chaturvedi P, Tyagi SC, Tyagi N. Curcumin-
primed exosomes mitigate endothelial cell dysfunction during hyperho-
mocysteinemia. Life Sci 2014;107(1–2):1-7.
7. Simpson RJ, Jensen SS, Lim JW. Proteomic profiling of exosomes:
current perspectives. Proteomics 2008;8(19):4083-99.
8. Mathivanan S, Ji H, Simpson RJ. Exosomes: extracellular organelles important
in intercellular communication. J Proteomics 2010;73(10):1907-20.
9. Alvarez-Erviti L, Seow Y, Yin H, Betts C, Lakhal S, Wood MJ. Delivery
of siRNA to the mouse brain by systemic injection of targeted exosomes.
Nat Biotechnol 2011;29(4):341-5.
10. Wahlgren J, De LKT, Brisslert M, Vaziri Sani F, Telemo E, Sunnerhagen
P, et al. Plasma exosomes can deliver exogenous short interfering RNA
to monocytes and lymphocytes. Nucleic Acids Res 2012;40(17)e130.
11. Pan Q, Ramakrishnaiah V, Henry S, Fouraschen S, de Ruiter PE,
Kwekkeboom J, et al. Hepatic cell-to-cell transmission of small silencing
RNA can extend the therapeutic reach of RNA interference (RNAi). Gut
2012;61(9):1330-9.
12. Shtam TA, Kovalev RA, Varfolomeeva EY, Makarov EM, Kil YV,
Filatov MV. Exosomes are natural carriers of exogenous siRNA to
human cells in vitro. Cell Commun Signal 2013;11:88, https://doi.org/
10.1186/1478-811X-11-88.
13. Johnsen KB, Gudbergsson JM, Skov MN, Pilgaard L, Moos T, Duroux
M. A comprehensive overview of exosomes as drug delivery vehicles -
endogenous nanocarriers for targeted cancer therapy. Biochim Biophys
Acta 2014;1846(1):75-87.
14. Chen L, Charrier A, Zhou Y, Chen R, Yu B, Agarwal K, et al. Epigenetic
regulation of connective tissue growth factor by MicroRNA-214 delivery
in exosomes from mouse or human hepatic stellate cells. Hepatology
2014;59(3):1118-29.
15. Haney MJ, Klyachko NL, Zhao Y, Gupta R, Plotnikova EG, He Z, et al.
Exosomes as drug delivery vehicles for Parkinson's disease therapy. J
Control Release 2015;207:18-30.
16. Zhuang X, Xiang X, Grizzle W, Sun D, Zhang S, Axtell RC, et al.
Treatment of brain inflammatory diseases by delivering exosome
encapsulated anti-inflammatory drugs from the nasal region to the
brain. Mol Ther 2011;19(10):1769-79.
17. Sun D, Zhuang X, Xiang X, Liu Y, Zhang S, Liu C, et al. A novel
nanoparticle drug delivery system: the anti-inflammatory activity of
curcumin is enhanced when encapsulated in exosomes. Mol Ther
2010;18(9):1606-14.
204 M.S. Kim et al / Nanomedicine: Nanotechnology, Biology, and Medicine 14 (2018) 195–204
18. Mulcahy LA, Pink RC, Carter DR. Routes and mechanisms of
extracellular vesicle uptake. J Extracell Vesicles 2014;3, https://
doi.org/10.3402/jev.v3.24641.
19. Montecalvo A, Larregina AT, Shufesky WJ, Stolz DB, Sullivan ML,
Karlsson JM, et al. Mechanism of transfer of functional microRNAs
between mouse dendritic cells via exosomes. Blood 2012;119(3):756-66.
20. Kaur S, Singh SP, Elkahloun AG, Wu W, Abu-Asab MS, Roberts DD.
CD47-dependent immunomodulatory and angiogenic activities of
extracellular vesicles produced by T cells. Matrix Biol 2014;37:49-59.
21. Long KB, Beatty GL. Harnessing the antitumor potential of macrophages
for cancer immunotherapy. Oncoimmunology 2013;2(12)e26860.
22. Kim MS, Haney MJ, Zhao Y, Mahajan V, Deygen I, Klyachko NL, et al.
Development of exosome-encapsulated paclitaxel to overcome MDR in
cancer cells. Nanomedicine 2016;12(3):655-64.
23. Yuan D, Zhao Y, Banks WA, Bullock KM, Haney M, Batrakova E, et al.
Macrophage exosomes as natural nanocarriers for protein delivery to
inflamed brain. Biomaterials 2017;142:1-12.
24. Vilner BJ, John CS, Bowen WD. Sigma-1 and sigma-2 receptors are
expressed in a wide variety of human and rodent tumor cell lines. Cancer
Res 1995;55(2):408-13.
25. Banerjee R, Tyagi P, Li S, Huang L. Anisamide-targeted stealth
liposomes: a potent carrier for targeting doxorubicin to human prostate
cancer cells. Cancer 2004;112(4):693-700.
26. Kim SK, Foote MB, Huang L. The targeted intracellular delivery of
cytochrome C protein to tumors using lipid-apolipoprotein nanoparti-
cles. Biomaterials 2012;33(15):3959-66.
27. Yang Y, Hu Y, Wang Y, Li J, Liu F, Huang L. Nanoparticle delivery of
pooled siRNA for effective treatment of non-small cell lung cancer. Mol
Pharm 2012;9(8):2280-9.
28. Kooijmans SA, Fliervoet LA, van der Meel R, Fens MH, Heijnen HF,
van Bergen En Henegouwen PM, et al. PEGylated and targeted
extracellular vesicles display enhanced cell specificity and circulation
time. J Control Release 2016;224:77-85.
29. Sena-Esteves M, Tebbets JC, Steffens S, Crombleholme T, Flake AW.
Optimized large-scale production of high titer lentivirus vector
pseudotypes. J Virol Methods 2004;122(2):131-9.
30. Bolte S, Cordelieres FP. A guided tour into subcellular colocalization
analysis in light microscopy. J Microsc 2006;224(Pt 3):213-32.
31. Maruo Y, Gochi A, Kaihara A, Shimamura H, Yamada T, Tanaka N, et al.
ICAM-1 expression and the soluble ICAM-1 level for evaluating the
metastatic potential of gastric cancer. Cancer 2002;100(4):486-90.
32. Cullen R, Germanov E, Shimaoka T, Johnston B. Enhanced
tumor metastasis in response to blockade of the chemokine receptor
CXCR6 is overcome by NKT cell activation. J Immunol
2009;183(9):5807-15.
33. Peer D, Karp JM, Hong S, Farokhzad OC, Margalit R, Langer R.
Nanocarriers as an emerging platform for cancer therapy. Nat
Nanotechnol 2007;2(12):751-60.
34. Davis ME, Zuckerman JE, Choi CH, Seligson D, Tolcher A, Alabi CA,
et al. Evidence of RNAi in humans from systemically administered
siRNA via targeted nanoparticles. Nature 2010;464(7291):1067-70.
35. Batrakova EV, Li S, Brynskikh AM, Sharma AK, Li Y, Boska M, et al.
Effects of pluronic and doxorubicin on drug uptake, cellular metabolism,
apoptosis and tumor inhibition in animal models of MDR cancers. J
Control Release 2010;143(3):290-301.
36. Lotvall J, Hill AF, Hochberg F, Buzas EI, Di Vizio D, Gardiner C, et al.
Minimal experimental requirements for definition of extracellular
vesicles and their functions: a position statement from the International
Society for Extracellular Vesicles. J Extracell Vesicles 2014;3, https://
doi.org/10.3402/jev.v3.26913.
37. Finn OJ. Immuno-oncology: understanding the function and dysfunction
of the immune system in cancer. Ann Oncol 2012;23(Suppl 8):viii6-9.
38. Parolini I, Federici C, Raggi C, Lugini L, Palleschi S, De Milito A, et al.
Microenvironmental pH is a key factor for exosome traffic in tumor cells.
J Biol Chem 2009;284(49):34211-22.
39. Morishita M, Takahashi Y, Nishikawa M, Takakura Y. Pharmacokinet-
ics of exosomes — an important factor for elucidating the biological
roles of exosomes and for the development of exosome-based
therapeutics. J Pharm Sci 2017;106(9):2265-9.
40. Kamerkar S, LeBleu VS, Sugimoto H, Yang S, Ruivo CF, Melo SA, et al.
Exosomes facilitate therapeutic targeting of oncogenic KRAS in pancreatic
cancer. Nature 2017;546(7659):498-503.