In vitro and in vivo RNA inhibition by CD9-HuR functionalized

Subscriber access provided by Columbia University Libraries

Communication

In vitro and in vivo RNA inhibition by CD9-HuR functionalized exosomes encapsulated with miRNA or CRISPR/dCas9 Zhelong Li, Xueying Zhou, Mengying Wei, Xiaotong Gao, Lianbi Zhao, Ruijing Shi, Wenqi Sun, Yunyou Duan, Guodong Yang, and Lijun Yuan Nano Lett., Just Accepted Manuscript • DOI: 10.1021/acs.nanolett.8b02689 • Publication Date (Web): 05 Dec 2018 Downloaded from http://pubs.acs.org on December 5, 2018

Just Accepted “Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted online prior to technical editing, formatting for publication and author proofing. The American Chemical Society provides “Just Accepted” as a service to the research community to expedite the dissemination of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully peer reviewed, but should not be considered the official version of record. They are citable by the Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore, the “Just Accepted” Web site may not include all articles that will be published in the journal. After a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web site and published as an ASAP article. Note that technical editing may introduce minor changes to the manuscript text and/or graphics which could affect content, and all legal disclaimers and ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or consequences arising from the use of information contained in these “Just Accepted” manuscripts.

is published by the American Chemical Society. 1155 Sixteenth Street N.W., Washington, DC 20036 Published by American Chemical Society. Copyright © American Chemical Society. However, no copyright claim is made to original U.S. Government works, or works produced by employees of any Commonwealth realm Crown government in the course of their duties.

Page 1 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

In vitro and in vivo RNA inhibition by CD9-HuR functionalized exosomes encapsulated with miRNA or CRISPR/dCas9

Zhelong Li1,2,#, Xueying Zhou1,2,#, Mengying Wei2,#, Xiaotong Gao2,3, Lianbi Zhao1,2, Ruijing Shi1,2, Wenqi Sun1,2, Yunyou Duan1, Guodong Yang2,*, Lijun Yuan1,*

1Department

of Ultrasound Diagnostics, Tangdu Hospital, Fourth Military Medical

University, Xi’an, 710038, People’s Republic of China 2The

State Laboratory of Cancer Biology, Department of Biochemistry and Molecular

Biology, Fourth Military Medical University, Xi’an, 710032, People’s Republic of China 3Department

of Hematology, Tangdu Hospital, Fourth Military Medical University, Xi’an,

710038, People’s Republic of China # These authors contributed equally to this article.

ACS Paragon Plus Environment

Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

ABSTRACT In vitro and in vivo delivery of RNAs of interest holds promise for gene therapy. Recently, exosomes are considered as the rational vehicle for RNA delivery, especially miRNA and/or siRNA, while the loading efficiency is limited. In this study, we engineered the exosomes for RNA loading by constructing a fusion protein, in which the exosomal membrane protein CD9 was fused with RNA binding protein, while the RNAs of interest either natively harbors or is engineered to have the elements for the binding. By proof-of-principle experiments, we here fused CD9 with HuR, an RNA binding protein interacts with miR-155 with a relatively high affinity. In the exosome packaging cells, the fused CD9-HuR successfully enriches miR-155 into exosomes when miR-155 was excessively expressed. Moreover, miR-155 encapsulated in the exosomes in turn could be efficiently delivered into the recipient cells and recognizes the endogenous targets. In addition, we also revealed that the CD9-HuR exosomes could enrich the functional miRNA inhibitor or CRISPR/dCas9 when the RNAs were engineered to have the AU rich elements. Taken together, we here have established a novel strategy for RNA cargo encapsulation into engineered exosome, which in turn functions in the recipient cells.

Keywords: exosomes; miRNA; CRISPR/Cas9; RNA binding protein; gene engineering


Page 2 of 26

Page 3 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Introduction Therapeutically targeting aberrant gene expression holds promise for the disease treatment. miRNAs, the 19-26 nt non-coding regulatory RNAs, are found to play important

roles

in

development,

physiological

processes,

and

diseases,

via

posttranscriptionally regulating target mRNAs with the recognition sites1, 2. Besides miRNAs, long non-coding RNAs (lncRNAs) are recently found to be also important in gene regulation and thus biological processes3. It is thus very important to deliver the miRNAs, lncRNAs, and/or even functional mRNAs of interest, for both mechanism study and disease control. Intracellular delivery of RNAs via nanoparticles or virus vectors mediated RNA delivery is commonly used to intervene gene expression4-6. However, these nucleic acids delivery methods have robust toxic side effects. Exosomes, a kind of extracellular microvesicles with a diameter range of 30–150 nm, are emerging as a rational drug carrier7, 8. Exosomes carry different types of cargos, such as RNAs, proteins and small molecule drugs, and release these cargos in the recipient cell. For miRNA/siRNA encapsulation, the miRNA/siRNA of interest could be either loaded with electroporation, or encapsulated in the donor cells transfected with the target miRNA9, 10. However, these strategies have relatively poor encapsulation efficiency. In addition, there still remains no effective strategy for large RNA delivery into exosomes. The endogenous RNAs sorting out into exosomes are fine-tuned11,

12.

It is well

established that RNA binding proteins are involved in the RNA sorting out process13-15. In addition, exosomes have been recently engineered towards better targeting specificity via exosome surface protein modifications16. For example, tetraspanin CD9, an exosomal surface marker was commonly re-engineered for the purpose of tracking and target delivery17-19. In light of the exosome engineering strategies and mechanisms for RNA sorting out into exosomes, we hypothesized that fusion of the exosomal membrane protein with specific RNA binding protein would increase the loading efficiency of the RNA targets of interest. In this study, we fused CD9 with HuR (human antigen R), an RNA binding protein interacts with AREs in the target RNA with a relatively high


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

affinity20, and explored its efficiency in exosome-based loading and delivering miR-155 and Cas9 mRNA engineered with AREs. Our study here has established a novel strategy to load RNA cargos into exosomes, which might be promising in gene therapy.


Page 4 of 26

Page 5 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Results Surface functionalization of exosomes by CD9-HuR fusion protein Exosomes have been well studied as a drug delivery vehicle. Up to now, low cargo encapsulation efficiency remains the big hurdle for its potential application. In light of the selective sorting of the endogenous RNAs into exosomes by RNA binding proteins7, 13, we asked whether exosomes could be engineered to selectively sort RNA targets of interest. To this end, we explored the possibility of engineering of the tetraspanin CD9 with RNA binding protein HuR for a proof-of-principle study. As we know, both the Nand C-termini of CD9 localize in the intracellular side (Figure 1a). To minimize the influence of fusion protein on the transport of the membrane protein, RNA binding protein HuR was chosen to fuse to the C-terminus CD9 by gene engineering method (Figure 1b). The recombinant vector produced abundant fusion protein in the cells when transfected (Figure 1c, Figure S1). Moreover, the engineered CD9-HuR fusion protein could be also sorted into the exosomes (Figure 1c, Figure S1). To explore whether HuR localizes in the intracellular space of the exosome, CD9-HuR engineered exosomes were incubated with anti-HuR or anti-CD9 antibodies, followed by immunoprecipitation. No exosomes were successfully pulled down by anti-HuR, while abundant exosomes were pulled down by anti-CD9 (Figure 1d), suggesting the HuR localized in the inside of the exosome (Figure 1e).


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 1. Exosome surface functionalization with CD9-HuR fusion protein. (a) Schematic representation of membrane localization of the tetraspanin CD9. Both the N- and C-terminus localize in the intracellular side. (b) Schematic representation of RNA binding protein HuR fused to the C-termini of CD9. (c) Western blot analysis of the expression of fusion protein CD9-HuR in both the parental cells and derived exosomes with indicated treatments. GAPDH served as the loading control. Representative data of three different experiments. (d) Exosomes from CD9-HuR expressing cells were isolated and further pulled down by IgG, anti-HuR, or anti-CD9 antibodies. No exosomes were successfully pulled down, suggesting that HuR localized inside of the exosome. Representative image of three different experiments. (e) Schematic representation of the CD9-HuR functionalized exosome structure.

In order to explore whether surface functionalization by CD9-HuR affects the exosome size and morphology, the CD9-HuR modified exosomes were further characterized. Compared with the non-modified control exosomes, CD9-HuR functionalization didn’t change the marker expression, size, and morphology significantly (Figure 2a-c).


Page 6 of 26

Page 7 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 2. Characterization of the CD9-HuR fusion protein functionalized exosomes. (a) Western blot analysis of the inclusive exosome markers TSG101 and CD63, and the exclusive marker GM130. Representative data of three different experiments. (b) Size distribution of the non-modified exosomes and exosomes from CD9-HuR expressing vector transfected cells were analyzed by dynamic light scattering. Data represent measurements of 4 biological samples. (c) Representative electron microscope image of the control and CD9-HuR functionalized exosomes. Scale bar=100 nm.

Efficient loading of miR-155 into exosomes by CD9-HuR fusion protein In the following experiments, we explored whether the engineered CD9-HuR fusion protein could promote RNA of interest loading during the exosome biogenesis. As a proof-of-principle study, we first focused on the miRNAs for its easy manipulation. miR-155 and miR-328 were selected, as miR-155 is AU rich while miR-328 is AU poor (Figure 3a) 21. Either miR-155 or miR-328 was incubated with HuR and RNA-IP analysis revealed that HuR interacted with miR-155 rather than miR-328 (Figure 3b). Next, we asked whether CD9-HuR fusion protein could promote miR-155 encapsulation in the exosome packaging cells. HEK293T cells were co-transfected with 100 nM miR-155 and CD9-HuR expressing vector, or the corresponding control miRNA and vectors (Figure 3c). As revealed by qPCR, CD9-HuR greatly enriched miR-155 into the exosomes when miR-155 was overexpressed in the packaging cells. Correspondingly, miR-155 level in the parental cells was significantly reduced (Figure 3d, 3e). Absolute qRT-PCR data (by comparing with the cDNA synthesized from quantified miR-155 sample) revealed that there were estimated 13.9±6.2 copies of miR-155 per the control exosomes, while 98.2±13.6 copies per CD9-HuR modified exosomes. In contrast, CD9-HuR had no effects on the enrichment of miR-328 into the exosomes (Figure 3f, 3g).


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 3. Efficient miR-155 encapsulation into the CD9-HuR fusion protein functionalized exosomes. (a) Sequences of miR-155 and miR-328. (b) RNA-IP analysis of the interaction between miR-155/miR-328 and HuR. About 10 nM miRNAs were incubated with 1 µg HuR antibody for further RNA-IP analysis. Data are expressed as mean±SEM of three different experiments. *, p < 0.05. (c) Illustration of the procedure how miR-155 or miR-328 was encapsulated by the CD9-HuR fusion protein functionalized


Page 8 of 26

Page 9 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

exosomes. About 100 nM miR-155 or miR-328 was cotransfected into the HEK293T cells together with the control or CD9-HuR fusion expressing vector. Exosomes were isolated for further analyses. (d) Expression of miR-155 in HEK293T cells treated as indicated. (e) Expression of miR-155 in exosomes derived from HEK293T cells treated as indicated. (f) Expression of miR-328 in HEK293T cells treated as indicated. (g) Expression of miR-328 in exosomes derived from HEK293T cells treated as indicated. Data are expressed as mean±SEM of three different experiments. *, p < 0.05, **, p < 0.01; ****, p < 0.0001.

CD9-HuR fusion protein functionalized exosomes effectively deliver miR-155 into recipient cells In view of above data on enrichment of miR-155 by CD9-HuR, we asked whether miR-155 was permanently trapped by HuR or could be released when the exosomes were endocytosed by the recipient cells. Monocyte cell line THP1 was included as miR-155 has been found to act importantly in the monocytes. THP1 cells were incubated with control or CD9-HuR functionalized exosomes (Figure 4a) and exosome tracking analysis revealed that these exosomes could be efficiently endocytosed by the THP1 cells (Figure 4b). Moreover, miR-155 level was significantly increased when the cells received exosomes from miR-155 and CD9-HuR co-transfected donor cells (Figure 4c). Accordingly, SOCS1 (suppressor of cytokine signaling 1), the known downstream target of miR-15522, 23, was found to be suppressed in the cells receiving the same exosomes (Figure 4d). All of these data revealed that CD9-HuR fusion protein functionalized exosomes effectively delivered miR-155 into recipient cells. The release of miR-155 might be explained by the competition of mRNA targets with high affinity in the cell or the great decrease of miR-155 concentration when entering the recipient cells.


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Figure 4. CD9-HuR functionalized exosomes deliver miR-155 into the recipient cells. (a) Illustration of the procedure how THP1 cells were treated with CD9-HuR functionalized exosomes encapsulated with miR-155. (b) CD9-HuR functionalized exosomes were stained with DiO and endocytosis of exosomes was analyzed by immunofluorescence microscope. Nuclei were counterstained with Hoechst. Scale bar=5 µm. (c) Expression of miR-155 in THP1 cells treated as indicated. Data are expressed as mean±SEM of three different experiments. *, p < 0.05. (d) Expressions of SOCS1 in THP1 cells treated as indicated were analyzed by western blot. GAPDH served as the loading control. Representative data of three different experiments. Quantification data were shown in the lower panel.

To further confirm whether CD9-HuR functionalized exosomes could deliver miR-155 in vivo, distribution of DiR labeled exosomes were thus tracked. Both bioluminescence imaging and immunofluorescence microscope analysis revealed that CD9-HuR functionalized exosomes mainly distributed in the liver and spleen as expected (Figure 5a-5c). In addition, exosomes from miR-155 and CD9-HuR co-transfected donor cells (ExomiR-155+CD9-HuR) significantly increased the expression of miR-155 in liver and spleen


Page 10 of 26

Page 11 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

(Figure 5d). Accordingly, expression of miR-155 downstream target SOCS1 was found greatly reduced in the liver and spleen only by ExomiR-155+CD9-HuR injection, while remained unchanged in the heart, lung, and kidney (Figure 5e). Collectively, these data indicate that the current strategy at least could functionally deliver miRNAs of interest into liver and spleen.

Figure 5. CD9-HuR functionalized exosomes deliver miR-155 in vivo. (a-c) Exosomes from CD9-HuR infected mouse fibroblasts were stained with DiR and about 100 µg (at protein level) exosomes in 100 µL were injected via tail vein. Liver and spleen dominated localization of the engineered exosomes as revealed by bioluminescence imaging (a, b) and immunofluorescence microscope (c). Scale bar=20 µm. (c).


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 12 of 26

Representative images of the data from 3 mice. (d) Expression of miR-155 in different tissues from mice treated with exosomes as indicated. Data are expressed as mean±SEM of three different experiments. *, p < 0.05; **, p < 0.01; ***, p < 0.001. (e) Expression of miR-155 target SOCS1 in different tissues from mice treated with exosomes as indicated. Representative images of at least three different experiments.

CD9-HuR fusion protein functionalized exosomes effectively deliver modified CRISPR/Cas9 system Besides miRNAs, long RNAs, like lncRNAs, have been found to be encapsulated into exosomes recently. It is thus highly possible that the proposed strategy could also load and thus deliver hurdle RNAs. Since in vivo delivery of CRISPR/Cas9 system holds great promise for gene editing and expression regulation24, we next explored whether CD9-HuR

fusion

protein

functionalized

exosomes

could

effectively

deliver

CRISPR/Cas9 system. To simplify the examination, our on hand available CRISPR/dCas9 mediated gene inhibition was included (Figure 6a). CRISPR/dCas9 significantly reduced the C/ebpα endogenous expression in the presence of C/ebpα gRNA (Figure 6b, 6c). To improve the loading efficiency of dCas9 mRNA into CD9-HuR exosomes, 3 × AREs were cloned downstream of the dCas9 stop codon (Figure 6d). Insertion of the 3 × AREs didn’t change the dCas9 mRNA level in the cells. However, dCas9 with AREs was more efficiently encapsulated into CD9-HuR exosomes when compared with dCas9 without AREs as shown by decreased dCas9 in parental cells (Figure 6e). As revealed by absolute qPCR data (by comparing with quantified dCas9 plasmid), there were estimated 2.4±1.6 copies of Cas9 in about 100 CD9-HuR modified exosomes when no AREs were fused, while about 22.3±8.5 copies of Cas9 were found in 100 CD9-HuR exosomes when 3 × AREs were fused. Consistently, exosomes from the C/ebpα gRNA, CD9-HuR and dCas9-ARE co-expressing cells significantly reduced C/ebpα expression in the recipient adipogenic stem cells, while the exosomes from the C/ebpα gRNA, CD9-HuR and dCas9 co-expressing cells had no obvious effects (Figure 6f). Moreover, exosomes from the C/ebpα gRNA, CD9-HuR and dCas9-ARE co-expressing cells significantly reduced C/ebpα expression in the liver 48 hours after tail vein injection (Figure 6g).


Page 13 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 6. CD9-HuR functionalized exosomes deliver CRISPR/dCas9 system in vivo. (a) Illustration of the CRISPR/dCas9 mediated repression of C/ebpα. (b) Sequencing result of the C/ebpα gRNA vector construction. (c) CRISPR/dCas9 significantly decreased endogenous expression of C/ebpα in the presence of gRNA. (d) Schematic representation of the lentivirus expressing dCas9 with or without AREs following the coding sequence. (e) Expression of dCas9 mRNA in packaging cells treated as indicated and their derivative exosomes. Data are expressed as mean±SEM of three biological replicates. *, p < 0.05. (f) Exosome packaging cells were forced to express CD9-HuR, C/ebpα gRNA, in combination with either control, dCas9, or dCas9-AREs. Harvested exosomes were incubated with adipogenic stem cells respectively. The effects on inhibition of the C/ ebpα endogenous expression in the recipient cells were analyzed by qPCR. (g) Exosomes as above were injected in vivo via tail vein. Endogenous expression of C/ ebpα in the liver was analyzed by qPCR. Data are expressed as mean±SEM of more than three different experiments. *, p < 0.05, ***, p < 0.001.


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 14 of 26

CD9-HuR fusion protein functionalized exosomes effectively alleviate CCl4 induced liver injury To confirm the efficacy of CD9-HuR functionalized exosomes in liver disease model, CCL4 induced liver injury model was included (Figure 7a). Previous researches have confirmed that the pro-inflammatory miR-155 is a detrimental promoter of hepatitis and fibrosis25-27, suggesting that therapeutically delivering miR-155 inhibitor would be beneficial for hepatitis prevention. To improve the loading efficiency of miR-155 inhibitor and minimizing the possible immunogenicity, antimiR-155 fused with 3 × AREs (referred as antimiR-155-AREs thereafter) or the negative control NC was synthesized and transfected into mouse liver cell line AML12 cells, together with the control empty or CD9-HuR vectors. And the resulting exosomes were denoted as ExoEmpty+NC,

ExoEmpty+antimiR-155-AREs,

ExoCD9-HuR+NC

and

ExoCD9-HuR+antimiR-155-AREs

respectively, As expected, CD9-HuR functionalization significantly enriched more antimiR-155-AREs

in

the

exosomes

(Figure

S2a).

Compared

with

the

ExoEmpty+antimiR155-AREs, ExoCD9-HuR+antimiR155-AREs treatment induced a more obvious decrease of endogenous miR-155 in the liver (Figure S2b). Consistently, expression level of inflammatory and fibrogenic genes, such as Tnfα, Mcp1, and Col1a1, were significantly lower in ExoCD9-HuR+antimiR-155-AREs treated group, compared with that in control exosome treated groups (Figure 7b-d). Sirius red staining further revealed that ExoCD9-HuR+antimiR-155-AREs treatment prevented the CCL4 induced liver fibrosis more efficiently (Figure 7e). All of these data further confirmed the efficacy and possible clinical application of CD9-HuR functionalized exosomes for in vivo RNA delivery.


Page 15 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Figure 7. CD9-HuR fusion protein functionalized exosomes effectively alleviate CCl4 induced liver injury. (a) Schematic representation of the experimental procedure. (b-d) Endogenous expression of Tnfα (b), Mcp1 (c), and Col1a1 (d) in livers from mice receiving indicated treatments. Livers were harvested 24 hours after the last exosome injection. U6 or β-actin served as internal control respectively. Data are expressed as mean±SEM of three different experiments. *, p < 0.05; **, p < 0.01. (e) Sirius red staining of the livers from mice receiving indicated treatments. Mice were treated with control or CCL4 twice a week for 3 weeks. About 200 µg indicated exosomes (at protein level) were injected 24 hours after each CCl4 treatment. Livers were harvested 24 hours after the last exosome injection for Sirius red staining. Representative images of n=5 mice in each group. Scale bar=250µm.


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Taken together, our study here indicates that in the packaging cells, CD9-HuR fusion protein recruits the target miRNAs or mRNAs to the exosomes via the RNA-HuR recognition. Once endocytosed in the recipient cells, these exosomes would release the encapsulated miRNAs or mRNAs, which act as functionally in the recipient cells (Figure 8).

Figure 8. Schematic summarization of the study. In the packaging cells, CD9-HuR fusion protein recruits the target miRNAs or mRNAs to the exosomes via the RNA-HuR recognition. In the recipient cells, miRNAs or mRNAs of interest are released from the CD9-HuR exosomes and thus act as functional miRNAs or mRNAs.


Page 16 of 26

Page 17 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Discussion In this study, by fusing CD9 C terminus with HuR, we engineered a kind of novel exosome with great capacity to enrich RNAs of interest. In addition, the enriched RNAs in the engineered exosomes could be functional in the recipient cells. Together, the proposed strategy is promising to be applied for in vivo gene delivery in the future in the clinical settings. To our knowledge, this is the first study aims to better the encapsulation capacity of exosomes. Exosomes are the natural membrane vesicles, which are now increasingly recognized as potential vehicles to deliver miRNA, mRNA, protein in vivo. At present, the most common way to load nuclei acids into exosomes is through electroporation, or direct encapsulation in the donor cells 9, 28, 29. However, the loading efficiency is relatively low. Increasing the loading efficiency of exosomes is badly needed for the translational medicine. Recently, miRNAs and other RNAs were found to be sorted out into exosomes via RNA binding proteins and other unknown mechanisms. In light of these studies, we fused the RNA binding protein HuR with CD9 for a proof-of-principle study. In the fusion protein, HuR is localized in the inside of exosomes. The preliminary data here indicate that the fusion protein helps sorting HuR binding RNAs into the exosomes during exosome biogenesis, especially when the RNA of interest is exogenously overexpressed. Notably, there are hundreds of RNA binding proteins expressed by the cell, which bind different RNAs mainly via the corresponding RNA elements. To this end, it is interesting to test other RNA binding protein-CD9 combination in enriching certain targets. Alternatively, the RNA of interest could be engineered to harbor the consensus sequence, such as the AREs corresponding to HuR tested in the present study. The latter strategy is of great significance for loading lncRNAs and mRNAs of interest. The nearly 5000 nt length of Cas9 mRNA makes CRISPR/Cas9 difficult to be capsulated into exosomes by electroporation or other strategies30. The strategy proposed here suggests that engineering Cas9 mRNA with HuR recognition motif (AU rich element) should be efficient to load Cas9 with our system, which is promising in in vivo gene editing and expression regulation. It is also important to note that forced expression of the RNAs of interest in the CD9-HuR exosome packaging cells would theoretically minimize the


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Page 18 of 26

loading of other endogenous RNAs recognized by HuR, and thus minimizing the unwanted effects. Once endocytosed, the cargos inside the exosomes could be either released into the cytoplasm or degraded when the exosomes fused with the lysosomes31,

32.

For the

engineered exosomes we proposed here, the enriched miR-155 and dCas9 should be also released from HuR before it functions in the recipient cells. Our data indeed reveal that these miR-155 can be released and thus functions when the exosomes are endocytosed in the recipient cells. One explanation might be that, in the recipient cells, the miR-155 concentration is low, and there are other targets of high affinity with miR-155. Alternatively, the CD9-HuR is preferentially degraded in the recipient cells. Further characterization of the releasing mechanisms would shed light on the refinement of the mechanism. For mRNA delivery, release of the mRNA from the fusion protein might be not necessary, as HuR was found to enhance the mRNA stability and translation efficiency33. Together, these data indicate that the current strategy is efficient in deliver functional nucleic acids, although certain details on how it works remain elusive. In conclusion, we have established a novel strategy to enrich RNAs of interest, such as miRNAs, lncRNAs, and even mRNAs in the exosomes, which largely improves the encapsulating capacity of exosome for nucleic acids. Regarding the advantage of exosomes in targeted delivery, this method can be widely used in gene expression regulation and thus disease therapy. In the future, it is worthwhile to evaluate the therapeutic potency of the proposed CD9-HuR exosome system in a disease model, such as liver cancer.


Page 19 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

Materials and methods Cell culture HEK293T cells, mouse liver AML12 cells, and adipogenic stem cells (ADSCs) were cultured in DMEM medium (Logan, Utah, USA) containing 10% fetal bovine serum and 1% penicillin-streptomycin (Logan, Utah, USA). THP1 cells were maintaining in RPMI medium (Logan, Utah, USA) plus with 10% FBS and 1% antibiotics. Cells were changed with fresh medium every other day and grown at 37°C in a 5% CO2 atmosphere. Plasmid construction CD9 cDNA was amplified by PCR using primers flanking with corresponding enzyme sites and the amplicons were cloned into pcDNA 3.1(-), with the resultant correct plasmids designated as pcDNA 3.1(-)-CD9. Similarly, HuR cDNA was amplified by PCR and then either cloned directly into pcDNA 3.1(-) or fused to the C terminus of CD9 in the pcDNA 3.1(-)-CD9. The whole CDS of the fusion protein was subcloned into pWPI with Pac1 for the purpose of lentivirus packing. All the clones were confirmed by sequencing and the right clones were stored for following application. For the construction of the C/ebpα gRNA expression vector, gRNA targeting C/ebpα promoter region

was

designed

using

the

online

CRISPR

Design

Tool

(http://tools.genome-engineering.org). The synthesized paired oligo strands with corresponding cohesive end were diluted and annealed, followed by cloning into the lenti sgRNA backbone after BsmBI digestion. The ORF of dCas9-NLS (nuclear localization sequence) was amplified from the dCAS9-VP64_GFP (Plasmid #61422)34 and cloned into pWPI with Pac1 and BstB1 enzymes. The resultant correct clone was designated as pWPI- dCas9. For the insertion of the AREs, paired oligo strands with multiple AREs and BstB1 cohesive end were annealed, followed by cloning into the pWPI- dCas9 after BsmBI digestion. PCR primers used in this article are listed in Table S1. Lentivirus package and infection The lentivirus expressing CD9-HuR, dCas9, dCas9-AREs, or gRNA were respectively transfected into 293T cells together with another two plasmids psPAX2 and pMD2G with Lipofectamine2000 (Invitrogen). The three plasmids (with the molar ratio of 10:5:1)


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

dissolved in FBS-free DMEM medium were mixed with Lipofectamine2000 and incubated at room temperature for 20mins. Then, 293T cells at 70-80% confluence received the transfection, and the medium was changed with fresh 10% FBS-containing medium six hours later. Lentivirus particles were collected from the medium supernatant filtered through a 0.45-um filter (Millipore) 48 hours post transfection and stored at -80°C before use. Exosome isolation HEK293T cells or other exosome packaging cells (AML12) were transfected with control or CD9-HuR expressing vectors by Lipofectamine2000 or infected with corresponding viruses as indicated, respectively. Twenty-four hours later, cells were further cultured in the FBS free medium for another 24-36 hours, followed by exosome isolation. For miR-155, antimiR155-AREs loading experiments, 100 nM miR-155, antimiR155-AREs or control miR-328 was co-transfected by Lipofectamine2000. Exoquick-TCTM kit was chosen for exosome isolation in this study. Although proteins, and nucleic acid/protein complexes are possibly co-precipitated in this method, our preliminary data revealed that the remaining nucleic acid/protein complexes had minimal effects on the loading efficiency and in vivo function analysis in our study. For the isolation of exosomes, supernatants were centrifuged at 500 g for 10 mins to remove cells and then at 10,000 g for 20 mins to eliminate the residual cellular debris. The resulting supernatant was regularly filtered through 0.4 μm filters. The sample was used to precipitate exosomes with Exoquick-TCTM kit. After that, exosomes were re-suspended in PBS or DMEM and stored at -80°C. For the miR-155, antimiR155-AREs and Cas9 mRNA loading efficiency analysis, the supernatants were incubated with 10 μg/mL RNase A for 30 min prior to exosome isolation, as exosomal RNAs are resistant to biochemical degradation by RNase A. Exosome characterization The morphology of isolated exosomes was analyzed by electron microscopy. Briefly, the exosomes were added onto the grid. The exosomes were then stained (2% uranyl acetate) and imaged by the electron microscope (JEM-2000EX TEM, JEOL Ltd., Tokyo, Japan).


Page 20 of 26

Page 21 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

For size distribution analysis of the exosomes, isolated exosomes from different sources were uniformly diluted to 500 ng/ml and the size distribution was analyzed by Nanoplus. Exosome pulldown assay To test whether HuR localized inside or outside in the engineered exosomes, we performed exosome pull down assay using anti-HuR (sc5261, Santa Cruz) and anti-CD9 (ab223052, Abcam) antibody. Briefly, about 50 µl of Protein-A Sepharose beads in 0.2% BSA in 1ml PBS was first incubated with 5 µg anti-HuR, anti-CD9 or control IgG at 4℃ overnight. Then, the anti-HuR or anti-CD9 and Protein-A Sepharose beads mixture were centrifuged to remove the unbound antibody. Finally, the antibody bound Protein-A Sepharose beads were incubated with the exosomes, followed by precipitation. The immunoprecipated exosomes were then subjected to lysis buffer, followed by western blot assay. HuR/miR-155 pull-down assay To explore the putative interaction between miR-155 and HuR, HEK 293T cells were transfected with HuR, and the cell lysates were incubated control or HuR antibody. The HuR/miR-155 interaction complexes were pulled down by routine IP procedure and the unbound miR-155 was washed with PBS for three times. The immunoprecipiated miR-155 was extracted using TRIzol® reagent, followed by miRNA reverse transcription and quantity detection by qPCR. Western Blotting Total protein from the indicated cells or exosomes was extracted using RIPA Lysis Buffer (Beijing, China) at 4°C for 30 min. Protein concentration was determined by PierceTM BCA Protein Assay Kit (Thermo, USA). Proteins were then concentrated on SDS-PAGE (6%) and separated by SDS-PAGE (12%). The gel was then transferred to nitrocellulose filter membrane before blotting assay. After blocking with 3% BSA, membranes were subsequently incubated with primary antibodies, anti-HuR (sc5261, Santa Cruz), anti-SOCS1 (PRS3765, Sigma), anti-CD9 (ab92726, Abcam), anti-GM130 (sc71166, Santa Cruz), anti-TSG101 (ab83, Abcam), anti-CD63 (sc5275, Santa Cruz), and anti-GAPDH (D110016-0100, BBI life sciences). After washing three times in TBST,


Nano Letters 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

The membranes were incubated with horseradish peroxidase-conjugated secondary antibody at room temperature for 1hour. Reverse transcription-polymerase chain reaction Total RNA was extracted using TRIzol® reagent (Invitrogen, IN, USA). miRNA and mRNA were reversely transcribed by miRcute Plus miRNA First-strand cDNA Synthesis Kit (Tiangen, Beijing, China) and Transcriptor Reverse Transcriptase(Indianapolis, IN, USA), respectively, following the manufacturers’ protocol. qPCR reactions (20 μl) were performed by FastStart Essential DNA Green Master (Indianapolis, IN, USA). miRNA and mRNA gene expression were normalized to U6 levels and β-actin for comparison, respectively. The sequences of PCR primers were provided in Table S1. The exosomal miRNA and Cas9 abundance per exosome was analyzed by absolute quantification qPCR together with exosome quantification. miRNA and Cas9 abundance per exosome was estimated as the quantified RNA abundance divided by exosome counts. The standard sample for miR-155 was prepared from the cDNA synthesized from miR-155 (1µL at the concentration of 1 nM) in a 20 µL system. The standard sample for Cas9 was prepared from the Pac1 enzyme cut pWPI-dCas9 plasmid. Exosome in vivo injection and analysis Male C56BL/6 mice（8-10 weeks old, 22-25g）were used. All animal experiments were carried out under protocols approved by the Animal Care and Use Committee of Fourth Military Medical University. For in vivo tracking exosomes, purified exosomes with indicated modifications were labelled with fluorescent dye DiR at the final concentration of 8 µM (Invitrogen). Labelled exosomes were collected by ultracentrifugation after washed with PBS, and stored in PBS before use. Mice were injected with labeled exosomes via tail vein injection. Different tissues were harvested 4 hours after injection for bioluminescence imaging, tissue sectioning, and qPCR analysis of miR-155. For slice sectioning, tissues were fixed for 15 mins by 4% paraformaldehyde and stained with Hoechst (Invitrogen) for counterstaining of the cell nuclei. The whole process is conducted in dark. The fluorescence signal for the labeled exosomes and the blue nuclei were viewed by laser scanning confocal microscope (ECLIPSE Ti, Nikon, Tokyo, Japan).


Page 22 of 26

Page 23 of 26 1 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17 18 19 20 21 22 23 24 25 26 27 28 29 30 31 32 33 34 35 36 37 38 39 40 41 42 43 44 45 46 47 48 49 50 51 52 53 54 55 56 57 58 59 60

Nano Letters

For downstream target SOCS1 expression analysis by western blot, different tissues were harvested 48 hours after injection. CCL4 was used to induce mouse liver fibrosis. Briefly, male C57BL/6 mice were injected with CCL4 (0.6 ml/kg of body weight, intraperitoneally) diluted in corn oil. Liver injury was induced by twice injection of CCL4 per week for 1-3 weeks. For exosome mediated therapy, control or CD9-HuR functionalized were exosomes from mouse liver cell line AML12 were loaded with NC or antimiR-155-AREs and injected 24 hours after the every CCL4 injection. At the end of the experiment, mice were sacrificed. Liver tissue was collected. Sections of formalin-fixed livers were stained with Sirius red staining for fibrosis analysis. RNA extracted from fresh liver tissues was reverse transcribed for qPCR analysis. Statistical analysis Data are expressed as mean±SEM. Student t test was used for two group comparison, and one-way ANOVA was used to compare the differences between groups while multiple comparisons were performed by Tukey’s post hoc test (Graphpad Prism 7.0). P values of

In vitro and in vivo RNA inhibition by CD9-HuR functionalized

Recommend Documents