Development of exosome surface display technology in living human cells

Accepted Manuscript Development of Exosome Surface Display Technology in Living Human Cells Zachary Stickney, Joseph Losacco, Sophie McDevitt, Zhiwen Zhang, Biao Lu PII:

S0006-291X(16)30247-9

DOI:

10.1016/j.bbrc.2016.02.058

Reference:

YBBRC 35361

To appear in:

Biochemical and Biophysical Research Communications

Received Date: 8 February 2016 Accepted Date: 15 February 2016

Please cite this article as: Z. Stickney, J. Losacco, S. McDevitt, Z. Zhang, B. Lu, Development of Exosome Surface Display Technology in Living Human Cells, Biochemical and Biophysical Research Communications (2016), doi: 10.1016/j.bbrc.2016.02.058. This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT 1

Development of Exosome Surface Display Technology in Living Human Cells Zachary Stickney, Joseph Losacco, Sophie McDevitt, Zhiwen Zhang, Biao Lu§

§

SC

California, USA, CA95053

Corresponding author

M AN U

BL: [email protected]

Email addresses:

SM: [email protected]

AC C

BL: [email protected]

EP

ZZ: [email protected]

TE D

ZS: [email protected] JL: [email protected]

RI PT

Department of Bioengineering, Santa Clara University, 500 El Camino Real, Santa Clara,

ACCEPTED MANUSCRIPT 2

ABSTRACT

Surface display technology is an emerging key player in presenting functional proteins for

RI PT

targeted drug delivery and therapy. Although a number of technologies exist, a desirable

mammalian surface display system is lacking. Exosomes are extracellular vesicles that facilitate cell-cell communication and can be engineered as nano-shuttles for cell-specific delivery. In this

SC

study, we report the development of a novel exosome surface display technology by exploiting mammalian cell secreted nano-vesicles and their trans-membrane protein tetraspanins. By

M AN U

constructing a set of fluorescent reporters for both the inner and outer surface display on exosomes at two selected sites of tetraspanins, we demonstrated the successful exosomal display via gene transfection and monitoring fluorescence in vivo. We subsequently validated our system by demonstrating the expected intracellular partitioning of reporter protein into sub-cellular

TE D

compartments and secretion of exosomes from human HEK293 cells. Lastly, we established the stable engineered cells to harness the ability of this robust system for continuous production, secretion, and uptake of displayed exosomes with minimal impact on human cell biology. In

EP

sum, our work paved the way for potential applications of exosome, including exosome tracking

AC C

and imaging, targeted drug delivery, as well as exosome-mediated vaccine and therapy.

ACCEPTED MANUSCRIPT 3

Keywords

RI PT

Exosome; Surface display; CD63; CD9; CD81

Abbreviations

SC

CMV, cytomegalovirus promoter; EF1α Elongation Factor 1-alpha promoter; FBS, fetal bovine serum; GFP, green fluorescent protein; HEK293, Human embryonic kidney cell line; MVB,

AC C

EP

TE D

M AN U

multi-vesicle body; RFP, Red Fluorescent protein; Puro, Puromycin resistant gene

ACCEPTED MANUSCRIPT 4

1.

Introduction

Surface display is an important technology for studying protein-protein or protein-ligand

RI PT

interaction [1]. Owing to its role in protein engineering, a number of platforms have been created, including phage display, bacterial surface display, yeast surface display and liposome surface display [2-5]. Recently, the emerging applications of surface display in targeted drug

SC

delivery and therapeutics have being widely appreciated [6-8]. However, the existing platforms were developed based upon either non-mammalian organisms or in vitro systems, which limit

M AN U

their applications in humans. In this study, we aim to develop a mammalian surface display system that is human compatible. Considering the complexity of mammalian cells, we chose to display candidate proteins on a mammalian cell-derived nano-vesicles, termed exosomes. Exosomes are lipid-bilayer-enclosed extracellular vesicles that transport signaling proteins, nucleic acids, and lipids among cells [9-11]. They are actively secreted by almost all

TE D

types of cells, exist in body fluids, and circulate in the blood [12]. Although the biogenesis of exosomes remains unclear, they are believed to be derived from endosomal-lysomal

EP

compartments [13]. Exosomes are initially formed during the inward budding of late endosomes and subsequently stored inside of multi-vesicular body (MVB) before being released into the

AC C

extracellular space [14]. Thus, it is not surprising that members of the endosomal forming and sorting proteins (Rab5, Rab27 and Rab35), heat-shock proteins, and tetraspanins (CD9, CD63 and CD81) are enriched in exosomes [15]. Tetraspanins are a special class of surface proteins that transverse four times of exosome

membrane. Among them, CD63 is the most abundant and is considered a hallmark of exosomes [16]. These trans-membrane proteins contain both extra- and intra-vesicular domains making them most suitable to display molecules on the surface of exosomes. Previously, our group and

ACCEPTED MANUSCRIPT 5

others have fused CD63 to a fluorescent reporter (GFP) and successfully used this chimeric protein to track the genesis and secretion of exosomes, along with the tissue-penetration and cellular-uptake of exosomes in vivo and ex vivo [17-18]. Here we further explore the surface

RI PT

engineering of exosome in living human cells using CD63 as a scaffold. This study establishes the groundwork for exosome surface engineering via tetraspanin CD63 and its family members

AC C

EP

TE D

M AN U

SC

CD9 and CD81.

ACCEPTED MANUSCRIPT 6

2.

Materials& methods

2.1

Design and construction of tetraspanin fusion proteins

RI PT

The C-terminal fusion expression vector of human tetraspanin CD63 with GFP or RFP is configured 5’→ 3’ as following: a constitutive cytomegalovirus (CMV) promoter, the coding sequences of the humanCD63, an in frame GFP or RFP, and a polyadenylation (Poly-A) signal.

SC

To add an antibiotics selection marker, a constitutive promoter EF1α with a puromycin

resistance gene was incorporated (Fig. 2a). Similarly, two other sets of fusion proteins of CD9

M AN U

and CD81 were constructed (Fig. 2c). The construction was carried out with a combination of PCR amplification of individual fragments and subsequently joined together using a seamless cloning kit (System Biosciences, Inc., Mountain View, CA, USA). To display a RFP reporter on the outer surface of exosome, we inserted the coding sequences of RFP into the second loop of CD63 (Fig. 3a &b). Additionally, a GFP-Rab5a fusion protein was assembled to serve as an

TE D

endosome marker. All final constructs were confirmed by DNA sequencing (ELIM BIO, Hayward, California). The genetically encoded protein sequences and gene ID are provided in the supplement files (Supplementary Table1 and 2). Cell culture and transfection

EP

2.2

AC C

Human embryonic kidney cells (HEK293) were cultured and maintained in high glucose Dulbecco’s Minimal Essential Medium (DMEM) supplemented with 10% fetal bovine serum, 2 mM GlutaMax (Life Technologies, Grand Island, NY, USA), 100 U/ml penicillin, and 100 U/ml streptomycin. At ~80-90% confluency, cells were passaged with 0.25% trypsin-EDTA for dissociation. All transfections were performed in 6-well plates. At 40~60% confluency, cells were transfected using plasmid DNA (1~2 µg/well) with PureFection Transfection Reagent

ACCEPTED MANUSCRIPT 7

(System Biosciences, Inc.,, Mountain View, CA, USA) or FuGENE6 transfection reagent (Promega Corp. Madison, WI) according to the user manuals. 2.3

Drug selection and stable cell lines establishment

RI PT

After transfection for 48 hours, HEK293 cells were switched to a culture medium

containing 5 µg/ml puromycin. Stable cell lines were considered to be established when they remained GFP-positive and puromycin resistant after long period (> 8 weeks) culture. Stable cell

SC

lines were maintained in complete medium under puromycin pressure and then switched to puromycin-free medium for at least two passages before conducting any experiments. Exosome preparation and purification

M AN U

2. 4

Exosomes were obtained from the supernatant of cells as described previously with minor modification (Ref 16). Briefly, stable cells were grown until 70~80% confluency. The spent medium was then replaced by fresh growth medium, and after an additional 48-hour

TE D

culture, the conditioned medium was collected. Following initial centrifugation for 30 minutes at 3000g, 40C, the collected medium was mixed with ExoQuick-TC (System Biosciences, Inc.,, California, USA) and subjected to another centrifugation under the same condition. The resulting

EP

pellet (exosome) was either re-suspended in a phosphate buffer solution for further analysis, or stored at -800C for future use.

Nanoparticle tracking analysis

AC C

2.5

Exosomes isolated from the engineered and control cells were both subjected to

nanoparticle tracking analysis, using NS300 (Malvern Instruments Ltd) as reported (19, 20). As done in a typical experiment, the exosome samples were subject to nanoparticle tracking analysis after a 1:5 dilution. The GFP fluorescence mode was used for the specific detection of de novo

ACCEPTED MANUSCRIPT 8

labeled exosomes from the engineered stable cell lines. Data was collected from 3 experiments under the same monitoring conditions and presented as mean ± SD. 2.6

Exosome uptake assay

RI PT

Equal amounts (50 µg) of exosomes isolated from the culture medium of either

engineered or parental control cells were added to the cultured HEK293 cells at confluency of

before imaging [16]. 2.7

Live cell monitoring with microscope

SC

80%. After 48 hours of incubation, cells were washed twice with the prewarmed PBS buffer

M AN U

Live cells were monitored using a LEICA DMI3000B fluorescent microscope. Data collection and processing was performed with LAS 3.8 software. The same field was subject to imaging under different conditions such as phase contrast, GFP and/or RFP. Imaging was further processed and merged using Adobe Photoshop CS program to illustrate the relationships of GFP

TE D

and/or RFP positivity.

Results

3.1

Exosome surface engineering strategy and experiment design

EP

3.

One common feature of surface display systems is the use of certain native surface

AC C

proteins as molecular scaffolds. Individual organism, such as bacterial phage, yeast and liposome anchors at least one such a scaffold protein on its surface.

For exosomes, the protein is

tetraspanins (reference here). All tetraspanins share similar "M"-shape topology on exosomal surface (Ref 21, 22), including two short intra-vesicle termini, two extra-vesicle loops, a small intra-vesicle loop and four transmembrane domains (Fig. 1a). Tetraspanins are relatively small

ACCEPTED MANUSCRIPT 9

(200~350 aa), making them particularly desirable for exosome display via molecular engineering in mammalian cells. To achieve this goal, we constructed fluorescent reporters with both the inner and outer

RI PT

surface modification of tetraspanin (Fig. 1a). Using these fluorescent reporters we were able to monitor the incorporation of the tetraspanin scaffold correctly onto exosomes (Fig. 1b &1c). This system will also allow us: (1) to establish stable cell lines for long-term study of the genesis

SC

and secretion of exosomes in living mammalian cells; (2) to isolate and analyze the endogenous exosomes via fluorescent surface markers (Fig. 1d &1e); (3) to provide a novel platform for

M AN U

supporting a broad future applications such as exosome biogenesis, targeted drug delivery, exosome-based therapy, protein-protein interaction, molecular imaging, and vaccination, etc. (Fig. 1f) 3.2

Tetraspanin CD63 scaffold enables surface display of functional fluorescent proteins

TE D

to endocytic compartments

We first generated a cohort of 6 chimeric proteins to test whether tetraspanins are suitable for the surface display with cultured HEK293 cells. To display a special protein to the inner

EP

surface of an exosome, we initially fused fluorescent proteins at the C-terminus of CD63 (Fig. 2a). Because CD63 contains the intrinsic membrane localization signal, we expected to see the

AC C

green fluorescent light-up of endocytic vesicles/granules. As demonstrated in Fig. 2b, CD63 is able to efficiently direct the GFP or RFP to the endocytic compartment, evidenced by the punctuated granular fluorescence in the cytosol (Fig. 2b, yellow arrows in upper panel). In contrast, cytosolic expression of either GFP or RFP alone revealed an even, diffused accumulation of reporters in control cells, which is consistent with their cytosolic distributions of native GPF or RFP (Fig. 2b, lower panel).

ACCEPTED MANUSCRIPT 10

We next assessed whether other tetraspanins (CD9, CD81) could also be used to directly fuse reporters into the same intracellular compartment (Fig. 2c). As expected, CD9 and CD81fusions also present distinct and granular signal within cytosol (Fig. 2d), similar to the

RI PT

pattern of CD63. Thus, tetraspanins in general can serve as a robust molecular scaffold to display different proteins on exosomes in cultured HEK293 cells.

To further determine whether the displayed fluorescent markers localize to the proper

SC

endocytic compartments within living cells, we constructed an endosomal reporter by tagging GFP to a well-known endosomal protein, Rab5a (Ref 25). As expected, GFP-Rab5A localizes to

M AN U

endocytic compartments, which perfectly matched the positive sites in all three tetraspanin fusions (CD9-RFP, CD63-RFP & CD81-RFP) (Supplementary Fig. 1S). These results strongly support the notion that tetraspanins display proteins on the surface of exosomes that are normally situated at endocytic compartments before being released into extracellular space.

TE D

Together, our results suggest that tetraspanins can serve as a molecular scaffold to direct fused peptides to endocytic compartments in exosomes. The readily detectable fluorescent markers indicate that the displayed peptides are structurally and functionally preserved, and the

EP

system is robust and reliable. Since CD63 is the most abundant surface marker of exosomes, it

3.3

AC C

was chosen for the further development for outer surface display.

A-S site in loop 2 of CD63 enables outer surface display

The two termini of CD63 molecule and the small loop between transmembrane domain 2

and 3 are situated inside of the exosome, whereas the other two loops are located outside (Fig. 3a). Of the two outside loops, the larger one (103aa) is more exposed with most antibodies developed against this region. Accordingly, we chose this loop at the site between Alanine 133

ACCEPTED MANUSCRIPT 11

and Serine134 (we coined it "A-S site") for the surface display. This site is localized in an accessible region; away from a complicated mushroom-like structure formed by multiple

targeted cells with minimum steric hindrance [26].

RI PT

disulfide bonds. Such a design will allow the displayed proteins to easily interact with the

To investigate whether the displayed peptide has reserved function after insertion in this loop, we constructed the ruby-RFP-CD63 fusion protein using the A-S site. Instead of using a

SC

native CD63, we inserted RFP into our previously validated CD63-GFP and built a tri-fusion with dual fluorescent reporters (Fig. 3b). This will help us by using GFP as a positive control to

M AN U

monitor the correct exosomal anchoring, and simultaneously using RFP to test the remaining function when displayed in vivo. As shown in Fig. 3c &3d, the fluorescent signals lighted up vesicles or granules early (24h after transfection). Some signals appeared more or less concentrated at certain micro-domains of vesicles, suggesting the potential focal points for

TE D

inverted budding processes (yellow arrow). Longer incubation time (72 h after transfection) leads to more intense and distinct fluorescence signals, which localized to intra-cellular granules or multi-vesicular body (MVB)-like structures (Fig. 3e-g). The co-localization of displayed RFP

EP

and GFP support the notion that the molecular scaffold does not damage the overall structure of

3.4

AC C

the fusion proteins and preserves their functions.

Stable cells secrete engineered exosomes into culture medium with intact function

To confirm the observation in our transient experiments, we carried out long-term studies

by establishing stable cell lines. As shown in Fig. 4a-b, genetically transformed HEK293 cells exhibited distinct fluorescence granules in the cytoplasm. In some areas, aggregation of GFPpositive granules was apparent, suggesting the sites of MVB. These results confirm our previous

ACCEPTED MANUSCRIPT 12

observations from transient transfection studies and provide cellular factories for permanently producing surface displayed exosomes. Importantly, long-term survival of engineered HEK293 cells supports a notion that this novel system is suitable for both short and long term studies of

RI PT

exosome biogenesis.

To further examine whether the engineered exosomes can mature and ultimately release into extracellular space, we isolated exosomes from the conditioned medium for nanoparticle

SC

tracking analysis. While little or no fluorescence signals are found in the control samples (Fig. 4c), CD63-GFP-positive nanoparticles are abundant as recorded by video under identical

M AN U

conditions (Fig. 4d), indicating that they are truly surface displayed exosomes from engineered cells. These GFP-positive particles have two major peaks at ~86 nm and ~160 nm (Fig. 4e), which is consistent with reported sizes of exosomes. Other minor peaks (~240 nm& ~300 nm) are also present; suggesting larger sized particles. Additional smaller peaks (<~80 nm) are visible

TE D

albeit in small amounts (Fig. 4e). Although the heterogeneous nature of exosomes is not new, our data provide direct evidence that this heterogeneity may exist under certain conditions. Finally, we examined the internalization process of these engineered exosomes. Internalization

EP

of exosomes occurs frequently at recipient cells via a direct fusion with the plasma membrane or endocytosis. After applying 50 µg exosomal preparation to the culture medium of stable cells

AC C

and incubating for 48 hours, the GFP-positive particles appeared in the cytosol of recipient cells (Fig. 4f, right panel), suggesting the up-take of engineered exosomes. In contrast, the control cells without GFP-displayed exosomes showed no GFP fluorescent background (Fig. 4f, left panel). These results indicate that surface displayed exosomes are functional and can be taken up by recipient cells as reported.

ACCEPTED MANUSCRIPT 13

4.

Discussion We have designed and validated a mammalian surface display system via genetic fusion

RI PT

to an abundant exosome membrane protein CD63. Like other surface display systems, this new tool will enable a broad spectrum of protein engineering for biotechnology and pharmaceutical applications, but with improved bio-compatibility with human tissues. Importantly, exosome

SC

surface display is potentially superior for applications such as targeted drug delivery and therapy due to its favorable characteristics. First, exosomes are natural nano-vesicles of 30~200 nm in

M AN U

diameter with intrinsic biocompatibility and a high level of tissue-penetration ability [6, 7]. Second, exosomes have a great safety profile and can be stored for longer time without losing function due to innate special lipids which provide a strong yet flexible protection layer to enclosed drug cargos [27]. Third, they are natural shuttles using cell surface receptor-mediated pathways to deliver their cargos, giving native exosomes with certain levels of built-in specificity

TE D

[28, 29]. Critically, various surface anchoring proteins make them the best candidates for surface engineering to achieve better tissue-specificity and delivery efficacy. Our choice of tetraspanins

EP

as anchoring scaffolds enables both drug-packaging and tissue-targeting via simultaneously inner and outer surface engineering.

AC C

Successfully displaying a candidate protein onto the surface of exosomes requires a sound strategy and the ability to overcome a number of technical hurdles. The first key step was to identify a native surface protein to serve as anchoring scaffold. We utilized tetraspaninsCD9, CD63, and CD81, as they are the surface markers of all exosomes. Second, we focused on on the multi-trans-membrane configuration of CD63 and successfully identified two candidate sites for displaying the fusion proteins on both inner and outer surfaces of the exosome. Third, we validated our system by demonstrating the correct intracellular partitioning of engineered protein

ACCEPTED MANUSCRIPT 14

into the proper endosomal compartments and eventually secreted exosomes into the culture medium. Lastly, by establishing the stably engineered HEK293 cells, we demonstrate the ability of this robust system to continuously produce, secret and uptake displayed exosomes with

RI PT

minimal effects on normal cell biology.

The ability to specifically engineer nano-vesicle exosomes has far-reaching implications in basic and applied biomedical fields [30-33]. As demonstrated in this study, exosome surface

SC

displaying of fluorescent reporters provides an effective way for investigating biogenesis,

secretion and up-take of exosomes. It is conceivable that by presenting a molecule such as single

M AN U

chain variable fragment (ScFv) of antibody on the outer surface of exosomes, a targeted delivery system may be engineered. Similarly, by tagging a therapeutic protein on the inner surface of exosomes, one may devise novel exosome-based therapeutics with the use of this novel exosome

AC C

EP

TE D

surface display system.

ACCEPTED MANUSCRIPT 15

References P.S. Daugherty, Protein engineering with bacterial display, Curr. Opin. Struct. Biol. 17 (2007) 474–480. doi:10.1016/j.sbi.2007.07.004.

[2]

G.P. Smith, Filamentous fusion phage: novel expression vectors that display cloned antigens on the virion surface., Science. 228 (1985) 1315–1317. doi:10.1126/science.4001944.

[3]

E.T. Boder, K.D. Wittrup, Yeast surface display for screening combinatorial polypeptide libraries., Nat. Biotechnol. 15 (1997) 553–557. doi:10.1038/nbt0697-553.

[4]

R.J. Simpson, S.S. Jensen, J.W.E. Lim, Proteomic profiling of exosomes: Current perspectives, Proteomics. 8 (2008) 4083–4099. doi:10.1002/pmic.200800109.

[5]

S. Fujii, T. Matsuura, T. Sunami, T. Nishikawa, Y. Kazuta, T. Yomo, Liposome display for in vitro selection and evolution of membrane proteins., Nat. Protoc. 9 (2014) 1578–91. doi:10.1038/nprot.2014.107.

[6]

L. Alvarez-Erviti, Y. Seow, H. Yin, C. Betts, S. Lakhal, M.J. a Wood, Delivery of siRNA to the mouse brain by systemic injection of targeted exosomes., Nat. Biotechnol. 29 (2011) 341–5. doi:10.1038/nbt.1807.

[7]

S. El-Andaloussi, Y. Lee, S. Lakhal-Littleton, J. Li, Y. Seow, C. Gardiner, et al., Exosome-mediated delivery of siRNA in vitro and in vivo., Nat. Protoc. 7 (2012) 2112– 26. doi:10.1038/nprot.2012.131.

[8]

S. Kuate, J. Cinatl, H.W. Doerr, K. Uberla, Exosomal vaccines containing the S protein of the SARS coronavirus induce high levels of neutralizing antibodies., Virology. 362 (2007) 26–37. doi:10.1016/j.virol.2006.12.011.

[9]

M. Mittelbrunn, F. Sánchez-Madrid, Intercellular communication: diverse structures for exchange of genetic information, Nat. Rev. Mol. Cell Biol. 13 (2012) 328–335. doi:10.1038/nrm3335.

[11]

SC

M AN U

TE D

EP

AC C

[10]

RI PT

[1]

E. Cocucci, G. Racchetti, J. Meldolesi, Shedding microvesicles: artefacts no more., Trends Cell Biol. 19 (2009) 43–51. doi:10.1016/j.tcb.2008.11.003.

H. Valadi, K. Ekström, A. Bossios, M. Sjöstrand, J.J. Lee, J.O. Lötvall, Exosomemediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells., Nat. Cell Biol. 9 (2007) 654–659. doi:10.1038/ncb1596.

[12 ] C. Théry, Exosomes: secreted vesicles and intercellular communications., F1000 Biol. Rep. 3 (2011) 15. doi:10.3410/B3-15.

ACCEPTED MANUSCRIPT 16

M. Record, K. Carayon, M. Poirot, S. Silvente-Poirot, Exosomes as new vesicular lipid transporters involved in cell–cell communication and various pathophysiologies, Biochim. Biophys. Acta - Mol. Cell Biol. Lipids. 1841 (2014) 108–120. doi:10.1016/j.bbalip.2013.10.004.

[14]

D.S. Choi, D.K. Kim, Y.K. Kim, Y.S. Gho, Proteomics, transcriptomics and lipidomics of exosomes and ectosomes, Proteomics. 13 (2013) 1554–1571. doi:10.1002/pmic.201200329.

[15]

S. EL Andaloussi, I. Mäger, X.O. Breakefield, M.J.A. Wood, Extracellular vesicles: biology and emerging therapeutic opportunities., Nat. Rev. Drug Discov. 12 (2013) 347– 57. doi:10.1038/nrd3978.

[16]

M.F. Peterson, N. Otoc, J.K. Sethi, A. Gupta, T.J. Antes, Integrated systems for exosome investigation, Methods. (2015). doi:10.1016/j.ymeth.2015.04.015.

[17]

M. Fong, J. Lesnik, G. Li, T. Antes, B. Lu, Cyto-TracersTM: Novel lentiviral-based molecular imaging tools, Biotechniques. 49 (2010) 839–840. doi:10.2144/000113553.

[18]

A. Suetsugu, K. Honma, S. Saji, H. Moriwaki, T. Ochiya, R.M. Hoffman, Imaging exosome transfer from breast cancer cells to stroma at metastatic sites in orthotopic nudemouse models, Adv. Drug Deliv. Rev. 65 (2013) 383–390. doi:10.1016/j.addr.2012.08.007.

[19]

C.Y. Soo, Y. Song, Y. Zheng, E.C. Campbell, A.C. Riches, F. Gunn-Moore, et al., Nanoparticle tracking analysis monitors microvesicle and exosome secretion from immune cells, Immunology. 136 (2012) 192–197. doi:10.1111/j.1365-2567.2012.03569.x.

[20]

S. a. Melo, L.B. Luecke, C. Kahlert, A.F. Fernandez, S.T. Gammon, J. Kaye, et al., Glypican-1 identifies cancer exosomes and detects early pancreatic cancer, Nature. 523 (2015) 177–182. doi:10.1038/nature14581.

[21]

N. Hassuna, P.N. Monk, G.W. Moseley, L.J. Partridge, Strategies for targeting tetraspanin proteins: Potential therapeutic applications in microbial infections, BioDrugs. 23 (2009) 341–359. doi:10.2165/11315650-000000000-00000.

[23]

SC

M AN U

TE D

EP

AC C

[22]

RI PT

[13]

M. Seigneuret, Complete predicted three-dimensional structure of the facilitator transmembrane protein and hepatitis C virus receptor CD81: conserved and variable structural domains in the tetraspanin superfamily., Biophys. J. 90 (2006) 212–227. doi:10.1529/biophysj.105.069666. D.D. Taylor, C. Gercel-Taylor, Exosomes/microvesicles: mediators of cancer-associated immunosuppressive microenvironments, Semin. Immunopathol. (2011) 1–14. doi:10.1007/s00281-010-0234-8.

ACCEPTED MANUSCRIPT 17

C. Théry, L. Zitvogel, S. Amigorena, Exosomes: composition, biogenesis and function., Nat. Rev. Immunol. 2 (2002) 569–579. doi:10.1038/nri855.

[25]

Y. Wang, L. Zhang, Y. Li, L. Chen, X. Wang, W. Guo, et al., Exosomes/microvesicles from induced pluripotent stem cells deliver cardioprotective miRNAs and prevent cardiomyocyte apoptosis in the ischemic myocardium., Int. J. Cardiol. 192 (2015) 61–69. doi:10.1016/j.ijcard.2015.05.020.

[26]

S.H. Petersen, E. Odintsova, T. a. Haigh, A.B. Rickinson, G.S. Taylor, F. Berditchevski, The role of tetraspanin CD63 in antigen presentation via MHC class II, Eur. J. Immunol. 41 (2011) 2556–2561. doi:10.1002/eji.201141438.

[27]

S. a a Kooijmans, P. Vader, S.M. van Dommelen, W.W. van Solinge, R.M. Schiffelers, Exosome mimetics: A novel class of drug delivery systems, Int. J. Nanomedicine. 7 (2012) 1525–1541. doi:10.2147/IJN.S29661.

SC

RI PT

[24]

M AN U

[28] D.M. Pegtel, K. Cosmopoulos, D. a Thorley-Lawson, M. a J. van Eijndhoven, E.S. Hopmans, J.L. Lindenberg, et al., Functional delivery of viral miRNAs via exosomes., Proc. Natl. Acad. Sci. U. S. A. 107 (2010) 6328–6333. doi:10.1073/pnas.0914843107. A. Hoshino, B. Costa-Silva, T.-L. Shen, G. Rodrigues, A. Hashimoto, M. Tesic Mark, et al., Tumour exosome integrins determine organotropic metastasis, Nature. 527 (2015) 329–335. doi:10.1038/nature15756.

[30]

B.K. Thakur, H. Zhang, A. Becker, I. Matei, Y. Huang, B. Costa-Silva, et al., Doublestranded DNA in exosomes: a novel biomarker in cancer detection., Cell Res. 24 (2014) 766–9. doi:10.1038/cr.2014.44.

[31]

C. Tetta, E. Ghigo, L. Silengo, M.C. Deregibus, G. Camussi, Extracellular vesicles as an emerging mechanism of cell-to-cell communication, Endocrine. 44 (2013) 11–19. doi:10.1007/s12020-012-*9839-0.

[32]

H. Valadi, K. Ekström, A. Bossios, M. Sjöstrand, J.J. Lee, J.O. Lötvall, Exosomemediated transfer of mRNAs and microRNAs is a novel mechanism of genetic exchange between cells., Nat. Cell Biol. 9 (2007) 654–659. doi:10.1038/ncb1596.

EP

AC C

[33]

TE D

[29]

T.N. Lamichhane, S. Sokic, J.S. Schardt, R.S. Raiker, J.W. Lin, S.M. Jay, Emerging roles for extracellular vesicles in tissue engineering and regenerative medicine., Tissue Eng. Part B. Rev. 21 (2014) 45–54. doi:10.1089/ten.TEB.2014.0300.

ACCEPTED MANUSCRIPT 18

Figure Legends Fig.1 System design of surface engineering of exosome. (a) Schematic of Surface

RI PT

engineering of exosome via tetraspanin proteins. CD63 is a "M" shape transmembrane protein with two termini and small middle loop on inner surface while two other loops exist on the outer surface of the exosome. Peptides can be fused with CD63 either at the second loop (outer surface

SC

display; red oval) or the C-terminus (inner surface display; green oval). (b) Delivery of CD63 fusion genes into living cells will partition the displayed fluorescent marker (green) on the

M AN U

surface of exosome via CD63 anchoring. (c & d)Engineered exosomes are released into the culture medium that can be recovered with polymer-based precipitation solution and centrifugation. (e) Enriched exosomes can be used for analysis or characterization. (f) A list of

TE D

potential applications.

Fig.2 Expression of tetraspanin chimeric proteins in mammalian cells. (a) Mammalian expression vector and the configuration of CD63-GFP. Promoter sequences derived from

EP

cytomegalovirus (CMV) or elongation factor 1 alpha (EF1a) are used to drive expression of the fusion and puromycin resistant genes respectively. (b) Exosome surface display of fluorescent

AC C

markers CD63 in living HEK293 cells, 48 hours after transfection. The upper panel shows the punctuate distribution of fluorescent signals (green/red), indicating endosomal-lysosomal compartment distribution of the CD63 fused reporters (yellow arrows). In contrast, the nonfusion GFP or RFP (as control) reveals an even cytoplasm distribution (lower panel). (c)Mammalian expression vectors of CD9-GFP/RFP and CD81-GFP/RFP respectively. (d) Exosome surface display of fluorescent markers (CD9 & CD81) in HEK293 cells.

ACCEPTED MANUSCRIPT 19

Fig. 3 Surface display using CD63 molecular scaffold. (a) Configuration of CD63 scaffold docking RFP marker to the second extra-vesicular loop. (b)Construction of expression vectors

RI PT

for CD63 fusion protein with fluorescent markers on the surface of exosomes. (c, d)Fluorescent markers (yellow arrows) localized to intracellular vesicles of living HEK293 cell (24 hours after transfection). (e, f, g) The co-localization of displayed RFP and GFP in living HEK293 cells (72

SC

hours after transfection).

M AN U

Fig. 4 Engineered stable cell lines express CD63-GFP and secrete surface displayed exosomes that can be taken up by recipient cells. Following the establishment of stable cell lines, images were taken from the same field to show GFP-positive cells (a) and the intracellular localization of GFP-positive vesicles (b) at 20× magnification. The screen-shots of NS300 of the

TE D

GFP-negative control cells (c) and GFP-positive (white arrows) of the engineered stable cells (d). The nanoparticle tracking analysis of GFP-positive vesicles released from the engineered stable cells (e). The superimposed red line indicates mean ± SD from 3 experiments. Two

EP

dominant peaks were recorded at ~160 nm and ~86 nm, along with minor peaks at ~284 nm, ~337 nm or <~80 nm. The concentration of the fluorescent particles used for the analysis is ~5 ×

AC C

106/ml. (f) CD63-GFP-positive nanoparticles are shown in cells supplied with engineered exosomes (right panel), versus the control without engineered exosomes (left panel), indicating exosomes uptake by recipient cells.

Supplemental Fig. 1S Co-localization of the endosomal marker Rab5a-GFP with 3 tetraspanin RFP fusions Images are taken 48 hours after co-transfection of endosomal tracer

ACCEPTED MANUSCRIPT 20

RAB5a-GFP and either of CD9, CD81, or CD63 RFP fusion plasmid in HEK293 cells at 20× magnification. The co-localization of endosomal tracer (green) with tetraspanins (red) indicates the displayed fluorescent proteins on the surface of exosomes are situated at the endocytic

AC C

EP

TE D

M AN U

SC

RI PT

compartments of cells.

ACCEPTED MANUSCRIPT 21

Acknowledgment This work was supported by a startup fund of the Department of Engineering, and funds from the

RI PT

Willem P. Roelandts and Maria Constantino-Reolandts Grant Program at Santa Clara University.

Conflict of interest

AC C

EP

TE D

M AN U

SC

BL was a previous employee of SBI.

ACCEPTED MANUSCRIPT

(a) membrane

RI PT

RNA & Protein cargo

Tetraspanin CD9, CD63, CD81 Peptide on the outer surface Peptide on the inner surface

exosome

(b)

Exosome

exosome

Cell

cytosol exosome cell culture medium

centrifuge

AC C

(d)

EP

TE D

(c)

M AN U

nucleus

SC

Tetraspanin chimeric gene

engineered exosome

(e)

Nano-particle analysis

(f) Potential applications • Exosome track and image • Targeted drug delivery • Therapeutics • Protein-protein interaction • Vaccine

ACCEPTED MANUSCRIPT

(c) CD63

PuroR

EF1α

CMV

Ori

CD9/CD81

Amp

G/RFP EF1α

PuroR

Ori

M AN U

Amp

GFP/RFP

SC

CMV

RI PT

(a)

(d)

EP

CD63-RFP

CD9-GFP

CD81-GFP

CD9-RFP

CD81-RFP

AC C

CD63-GFP

TE D

(b)

GFP GFP

RFP

ACCEPTED MANUSCRIPT

(a)

Outer surface

Membrane bilayer

N-CD63

Exosome

RFP

GFP

C-CD63

Ori

Amp

(c)

EF1α

PuroR

M AN U

CMV

SC

exosome

RI PT

Inner surface

Inside exosome

(b)

(e)

C

(f)

EP

TE D

(d)

RFP

(g)

AC C RFP

GFP

D GFP

Merge

ACCEPTED MANUSCRIPT

(b)

(c)

Phase + GFP (d)

RI PT

160

(106)

Particle Tracking Analysis

2.0

SC

1.5 1.0

86

M AN U

GFP

(e) Concentration (particles)

(a)

0.5 0

0

284 337

Size (nm)

100 200 300 400 500 600 700 800 900 1000

Control

AC C

EP

TE D

(f)

CD63-GFP

Phase CD63-GFP

GFP Negative Control

Control

Phase CD63-GFP

GFP