Exosomes from HCV-infected hepatocytes induce the activation of primary human hepatic stellate cells.

Next, we extended our findings by examining primary human hepatic stellate cells. Primary HSC exposed to HCV-exo displayed significant upregulation of the profibrotic markers COL1A1/A2/3A1/4A1, TGF-β1, TIMP1, CTGF, matrix metalloproteinase 2 (MMP-2), and α-SMA (Fig. 2A and ​andB).B). Immunofluorescence analysis for α-SMA in exosome-treated cells showed a characteristic myofibroblast-like pattern of activated HSC and increased α-SMA expression (Fig. 2C). Further, an increase in α-SMA expression at the protein level in exosome-treated primary HSC was observed (Fig. 2D). The results suggest exosomes from HCV-infected hepatocytes promote primary human hepatic stellate cell activation similarly to immortalized LX-2 cells.

Open in a separate window

FIG 2

Exosomes derived from HCV-infected hepatocytes induce upregulation of profibrogenic markers in primary human hepatic stellate cells. (A and B) qRT-PCR analysis was performed for COL1A1 and COL1A2, COL3A1, COL4A1, TGF-β1, TIMP-1, CTGF, MMP-2, and α-SMA from RNA of primary HSC incubated with HCV-exo. (C) Immunofluorescence analysis of α-SMA protein (green) in control or HCV-exo-exposed primary HSC showing the phenotypic alteration to activated myofibroblast-like cells. Cell nuclei were stained with DAPI (blue). (D) Western blot analysis of α-SMA in primary HSC exposed to HCV-exo. (Right) Quantification of the α-SMA protein level was normalized with GAPDH. The values and SD represent the results of three independent experiments. Statistical significance was analyzed using the two-tailed Student t test. *, P < 0.05; ns, not significant.

MicroRNAs are present in exosomes derived from HCV-infected hepatocytes.

We next investigated the molecules that activate the HSC. MicroRNAs are carried through the exosomes and play a major role in cellular function. We examined the miRNA profile in HCV-exo. Clustering analysis revealed that exosomes from HCV-infected cells expressed distinct patterns of miRNAs compared with exosomes from the mock-treated cells (Fig. 3A). Selected miRNAs—miR-19a, miR-20a, miR-92, and miR-195 (Fig. 3A)—were validated by qRT-PCR and normalized with spiked-in Caenorhabditis elegans miR-39 (cel-miR-39). HCV-exo were significantly enriched in miR-19a, miR-20a, and miR-195 (Fig. 3B).

Open in a separate window

FIG 3

MicroRNA expression profiling of exosomes derived from HCV-infected hepatocytes. (A) miRNA expression profiling of exosomes isolated from control or HCV-infected hepatocytes using miScript miRNA PCR Array. The array data were analyzed using free Web-based software (http://pcrdataanalysis.sabiosciences.com/mirna/arrayanalysis.php). (B) Validation of differentially expressed miRNAs in HCV-exo by qRT-PCR. *, P < 0.05; **, P < 0.01. The error bars indicate SD.

We validated our in vitro findings in a small cohort of sera from healthy volunteers and non-HCV-infected and HCV-infected samples by determining the level of circulatory miR-19a. We observed that miR-19a expression was significantly upregulated in HCV-infected fibrotic patients compared to healthy volunteers and non-HCV-related liver disease patients with similar fibrosis grades (Fig. 4A). Higher expression of miR-19a in sera from late-fibrosis patients from cross-sectional patient samples was also observed (Fig. 4B). When we examined circulatory miRNA levels of miR-195 in sera of HCV-infected fibrotic patients, we did not observe a significant modulation in HCV fibrotic patients compared to healthy volunteers. We decided to characterize the function of miR-19a in HCV-mediated fibrosis.

Open in a separate window

FIG 4

Upregulation of serum miR-19a levels in HCV-infected patients with liver fibrosis. (A) Scatter plots of serum miR-19a levels in healthy volunteers (HV) (n = 15), non-HCV-associated liver fibrosis patients (non-HCV) (n = 20), and HCV-infected fibrosis patients (HCV) (n = 44). (B) Increased expression of serum miR-19a levels in HCV-infected fibrosis patients from the early fibrosis stage (F0 to F2) (n = 33) to late-stage fibrosis (F3 and F4) (n = 11). The horizontal lines indicate the median values per group. Fold regulation is expressed as real-time quantitative (RQ) values based on the 2^−ΔΔCT method. A Mann-Whitney U test (nonparametric) was used to determine statistical significance.

Exosomes from HCV-infected hepatocytes mediate the profibrogenic effect on hepatic stellate cells by shuttling miR-19a.

To gain insight into the exosome-mediated uptake of miR-19a in HSC, we incubated LX2 cells with control cells or HCV-exo at different time points and analyzed miR-19a expression by qRT-PCR. A significant upregulation of miR-19a in LX2 cells was observed within 2 to 3 h (Fig. 5A). Similarly, primary human hepatic stellate cells also showed internalization of exosomal miR-19a (Fig. 5B). To visualize the transfer of exosomal miR-19a into HSC, Cy3-labeled miR-19a or control miR (control) was transfected into Rep2a cells, and exosomes were isolated. When LX2 cells were incubated with labeled miR-19a-containing HCV-exo, red fluorescence of Cy3-miR-19a was detected in the cytoplasm of LX2 cells exposed to exosomes (Fig. 5C).

Open in a separate window

FIG 5

Uptake of exosomal miR-19a by human hepatic stellate (LX2) cells. (A) Time course analysis of exosomal miR-19a uptake in LX2 cells. LX2 cells were exposed to control or HCV-exo at the indicated time points and harvested for cellular RNA and qRT-PCR analysis of miR-19a. (B) Primary human HSC exposed to control or HCV-exo and processed to determine miR-19a expression as described for panel A. (C) Exosomes were isolated from HCV-infected cells transfected with Cy3-labeled control or Cy3-labeled miR-19a mimic and added to LX2 cells. Uptake of Cy3-labeled miR-19a was assessed by confocal microscopy. The arrows indicate cytoplasmic Cy3-labeled miR-19a (red) in LX2 cells. Nuclei were stained with DAPI (blue). *, P < 0.05. The error bars indicate SD.

To further verify the exosome-mediated miR-19a uptake in HSC, we depleted endogenous levels of miR-19a by transfecting miR-19a antagonist in LX2 cells and incubating for 30 h (Fig. 6A). We exposed the miR-19a-depleted LX2 cells to HCV-exo 30 h post-antagonist transfection and observed upregulation of miR-19a similar to that of exosome-exposed cells (Fig. 6B and ​andC).C). Together, our data suggested exosome-mediated shuttling of miR-19a into HSC. Given the key role of Rab27 in exosome release (13), we transfected Rep2a cells with Rab27a small interfering RNA (siRNA), and exosomes were isolated. HCV-infected cells were also treated with 10 or 20 μM spiroepoxide (exosome release inhibitor). Exosomes were isolated and incubated with LX2 cells for 3 h. As expected, HSC exposed to exosomes isolated from spiroepoxide-treated cells showed concentration-dependent inhibition of miR-19a uptake (Fig. 6D). Similarly, Rab27a knockdown cells showed no sign of miR-19a uptake (Fig. 6D). These results further confirmed the exosome-mediated uptake of miR-19a by HSC.

Open in a separate window

FIG 6

Inhibition of exosome release or exosome biogenesis affects miR-19a delivery in LX2 cells. (A) Depletion of endogenous miR-19a after antagonist transfection was confirmed by miR-19a-specific qRT-PCR. (B) The endogenous miR-19a level was depleted by transfecting LX2 cells with anti-miR-19a, and the cells were left untreated or exposed to HCV-exo. Expression of miR-19a was measured by qRT-PCR and normalized with endogenous U6. (C) LX2 cells were exposed to control or HCV-exo isolated from HCV-infected cells. The cells showed upregulation of the miR-19a level by qRT-PCR. (D) Exosomes isolated from HCV-infected cells with or without treatment with 10 and 20 μM spiroepoxide (exosome release inhibitor) and transfected with Rab27a siRNA. LX2 cells were exposed to the respective exosome preparations for 3 h, and miR-19a was analyzed by qRT-PCR. (E) Hepatic stellate (LX2) cells were transfected with miR-19a-3p mimic or control miR (25 nM each), and RNA was isolated. qRT-PCR analysis was performed using control and miR-19a-transfected cellular RNA with gene-specific probes and normalized with endogenous marker 18S RNA. The values represent the results of three independent experiments and SD. Statistical significance was analyzed using the two-tailed Student t test: *, P < 0.05.

In order to evaluate the profibrogenic role of miR-19a-3p encapsulated in HCV-exo, we transfected LX2 cells with control or miR-19a-3p mimic and analyzed the expression of profibrotic genes. LX2 cells transfected with miR-19a-3p mimic showed significant upregulation of TGF-β1, MMP-2, TIMP1, and interleukin 6 (IL-6) genes (Fig. 6E).

Exosome isolation.

Exosomes from cell culture supernatants were isolated and characterized as described previously (10). Briefly, virus-infected cells were washed twice with PBS and then supplemented with DMEM containing 2% exosome-depleted serum (System Biosciences) and incubated for 3 days. To inhibit exosome release, the cells were treated with N-SMase spiroepoxide inhibitor (Santa Cruz) as described previously (38). Culture supernatants were collected and processed for exosome isolation by the ExoQuick method (System Biosciences).

BODIPY labeling of exosomes and uptake.

BODIPY_493/503 solution (1 mM; Molecular Probes) was added in the culture medium of Rep2a cells, and the cells were cultured for 1 h at 37°C in a CO₂ incubator. The cells were washed three times with PBS, and fresh medium containing exosome-depleted FBS was added to the cells. After 24 h of incubation, the medium was collected and processed for exosome isolation. LX2 cells were seeded in 4-chambered slides (Nunc). LX2 cells were washed with serum-free medium, and the BODIPY-labeled exosomes were added to the LX2 cells for 3 h. The cells were washed with Hanks' balanced salt solution and labeled with Alexa Fluor 647-conjugated wheat germ agglutinin (Life Technologies) to stain the plasma membranes of the cells. The cells were fixed, stained with DAPI (4′,6-diamidino-2-phenylindole), and observed under confocal microscopy (Olympus FV1000).

RNA isolation and qRT-PCR.

Total cellular or exosomal RNA was isolated using TRIzol reagent (Invitrogen, CA). cDNA was synthesized by using the respective miRNA-specific or U6-specific primers with a TaqMan microRNA reverse transcription kit or random hexamers with Superscript III reverse transcriptase for gene expression studies. Real-time PCR was performed for quantitation by using TaqMan universal PCR master mix and 6-carboxyfluorescein (FAM)-minor groove binder (MGB) probes for miRNAs or for gene expression assays. U6 (assay number 001973) or 18S rRNA (assay number Hs03928985_g1) was used as an endogenous control for intracellular microRNA or gene expression, respectively, while synthetic spiked-in C. elegans miR-39 added to the exosomes prior to RNA extraction was used as an endogenous control for exosomal miRNA expression. The relative expression levels were normalized to the endogenous control by using the 2^−ΔΔCT formula (ΔΔC_T = ΔC_T of the sample − ΔC_T of the untreated control). RNA isolation from serum samples, miRNA profiling by miRNA Array, and exosomal HCV RNA determination were carried out as described previously (10, 25). For microRNA arrays, we isolated exosomes from cell supernatants from mock- and HCV-infected hepatocytes and performed exosomal miRNA profiling using miScript miRNA PCR Array and Qiagen miRNA Array Web-based software.

Western blotting and immunofluorescence.

Cell lysates were subjected to polyacrylamide gel electrophoresis and transferred onto a nitrocellulose membrane. The membranes were blocked with 5% nonfat dry milk and incubated with specific antibodies. The membrane was probed with antibodies to SOCS3, cyclin D1, phosphorylated STAT3 (pSTAT3), STAT3, pSmad3, Smad3 (Cell Signaling), and α-SMA (Santa Cruz). Proteins were detected by using enhanced chemiluminescence (ECL). The membrane was reprobed with actin or glyceraldehyde-3-phosphate dehydrogenase (GAPDH) as an internal control. The densitometric scanning of Western blots was performed with ImageJ software (NIH). For immunofluorescence, primary HSC were seeded in 4-well chamber slides and exposed to HCV-exo for 2 days. The cells were washed in phosphate-buffered saline (PBS), fixed with 3.7% formaldehyde for 20 min at room temperature, and blocked with 3% bovine serum albumin (BSA) for 1 h. The fixed cells were permeabilized with 0.2% Triton X-100 for 5 min at room temperature. The cells were incubated with α-SMA-specific mouse monoclonal antibody (Santa Cruz Biotechnology) overnight at 4°C, washed, and incubated with anti-mouse Ig conjugated with Alexa 488 antibody (Molecular Probes) for 1 h at room temperature. Finally, the cells were washed, and the nuclei were stained with DAPI and mounted for confocal microscopy (Olympus FV1000).

Shuttling assay for Cy3-labeled miRNA.

Control miR or miR-19a mimic (Invitrogen) was labeled with a Label It siRNA Tracker Cy3 kit that covalently attaches the Cy3 marker to siRNA (Mirus, Madison, WI, USA), following the manufacturer's instruction. This allowed tracking of fluorescence-labeled miR-19a. Huh 7.5 (Rep2a) cells harboring the HCV full-length genome were transfected with 10 nM control or Cy3-labeled miR-19a using Interferin transfection reagent (Polyplus). The cells were maintained for 24 h in medium containing exosome-depleted FBS. Exosomes were isolated and incubated on LX2 cells or primary human hepatic stellate cells for 4 h. The cells were washed three times with 1× PBS before fixation and processed for confocal imaging as discussed above.

We thank Scott Friedman for LX2 cells and Hengli Tang for Rep2a cells.

This work was supported by research grant R01 DK081817 from the National Institutes of Health.