Incorporation of MV-HLSC in hepatoma cells inhibits proliferation and induces apoptosis. (A): PKH26-labeled MV-HLSC, or MV-fibr, or MV-fibr-CD29 internalization in HepG2 detected by confocal microscopy in the presence or absence of blocking antibodies against α4 integrin, CD29, and CD44 and by fluorescence-activated cell sorting (FACS) analyses (filled curve: MV-HLSC internalization by HepG2). Kolmogrov-Smirnov statistical analyses of MVs versus Ctrl: p < .001. Three experiments were performed with similar results. Scale bar = 10 μm. (B): Analysis of HepG2 (black columns) and hepatocellular carcinoma cell (HCC [gray columns]) proliferation by BrdU internalization after 48 hours in the presence of vehicle alone (vehicle) or of different doses of MV-HLSC, MV-fibr, or MV-fibr-CTR, or MV-fibr-CD29, or MV-RNase. Proliferation was evaluated on normal human Hep (blue columns) in the presence of vehicle alone or of MV-HLSC. (C): Analysis of HepG2 (black columns) and HCC (gray columns) apoptosis by terminal dUTP nickend labeling assay (TUNEL) assay. Cells were incubated with vehicle, vincristine, or different doses of MV-HLSC, or MV-fibr, or MV-fibr-CTR, or MV-fibr-CD29, or MV-RNase. Results in (B) and (C) are expressed as mean ± SD of six different experiments. (B and C): *, p < .05 versus vehicle; ^§, p < .05 versus MV-HLSC treatments. (D): Apoptosis of HepG2 coincubated with MV-HLSC and vincristine or Doxo was evaluated. To analyze the antiapoptotic effect of MV-HLSC on normal human Hep (blue columns), apoptosis was induced by 5 mM GalN. Results are expressed as mean ± SD of six different experiments. *, p < .05 versus vehicle; ^§, p < .05 versus MV-HLSC, vincristine or Doxo alone; #, p < .05 versus Hep; +, p < .05 versus Hep+GalN. Abbreviations: BrdU, 5-bromo-2-deoxyuridine; Doxo, doxorubicin; GalN, D-galactosamine; Hep, hepatocytes; HLSC, human adult liver stem cell; MV, microvesicles; MV-HLSC, HSLC-derived MV; MV-fibr, MVs derived from fibroblasts; MV-fibr-CD29, MVs derived from transfected fibroblasts overexpressing the CD29; MV-RNase, RNase-inactivated MVs.

In vivo antitumor effect of MV-HLSC. (A): Tumor growth in SCID mice injected subcutaneously with 3 × 10⁶ HepG2 and after 1 week treated weekly with vehicle (♦, HepG2) or 100 μg of MV-HLSC (□), with MV-RNase (▴) or with CD29 blocking antibody-treated MV-HLSC (○, MV-CD29). (B): Tumor weights after excision from severe combined immunodeficiency (SCID) mice at sacrifice. Results are expressed as mean ± SD of 12 mice for experimental group. *, p < .05 versus HepG2; ^§, p < .05 versus MV-HLSC. (C): Representative micrographs of H&E staining of tumor sections from SCID mice treated with vehicle alone (HepG2) or with MV-HLSC, MV-RNase, and MV-CD29. Large areas of necrosis were visible in tumors treated with MV-HLSC. Scale bar = 50 μm. (D): Quantification of Proliferating Cell Nuclear Antigen (PCNA)-positive cells/hpf was calculated. (E): Quantification of apoptotic cells/hpf was calculated. Data are expressed as mean ± SD of 12 mice for experimental group. *, p < .05 versus HepG2; ^§, p < .05 versus MV-HLSC. Abbreviations: HLSC, human adult liver stem cells; hpf, high power field; MV, microvesicles; MV-HSLC, HSLC-derived MV; MV-RNase, RNase-inactivated MVs; MV-CD29, MVs pretreated with anti-CD29 blocking antibody.

MV-mediated transfer of miRNAs with potential antitumor activity to HepG2. (A): The expression of miR451, miR223, miR24, miR125b, miR31, miR21, miR122, and RNU48 was evaluated by quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) in HLSC (gray bars) and MV-HLSC (white bars) with respect to HepG2. Data were expressed as log of Rq, normalized to GAPDH mRNA and to one for HepG2. Some of these miRNAs were significantly enriched in MVs with respect to their relative human adult liver stem cell (HLSC) and were less expressed or not detectable in HepG2. Data are the mean ± SD of four experiments. *, p < .05 versus HepG2. (B): HepG2 were incubated with HLSC-derived MV (MV-HLSC) 6 hours, overnight (ON, 16 hours), and 24 hours (h), and the transfer of selected miRNAs was evaluated by qRT-PCR. The difference in C_t values between HepG2 treated with vehicle or stimulated with MV-HLSC was shown for each miRNA. The RNU48 and the miR410, that was not present in MVs, were used as controls. Data are the mean ± SD of four experiments. (C): Relative expression of miR451, miR223, and miR31 in MV-HLSC (white bar) with respect to MVs derived from fibroblasts (MV-fibr [gray bar]) evaluated by qRT-PCR. Data were normalized to GAPDH mRNA, to one for MV-fibr and expressed as mean ± SD. Three experiments were conducted with similar results. *, p < .05 versus MV-fibr. (D): Representative micrographs showing the in vivo internalization in tumors of MV-HLSC labeled with PKH26 and SYTO RNA that stain MV membrane and RNA content, respectively. Three experiments were performed with similar results. Abbreviations: miRNA, micro RNA; Rq, relative quantification.

Effect of MVs derived from DCR-kd HLSC on in vitro HepG2 apoptosis and in vivo tumor growth. (A): Quantitative reverse transcriptase polymerase chain reaction (qRT-PCR) analysis of DCR1 and AGO2 transcripts in DCR-Kd (white bar) and CTR-A (gray bar). Western blot of Dicer protein expression (lower panel). Actin was used as endogenous control. *, p < .05 versus CTR-A. (B): Analysis of miRNAs in MV DCR− (white bar) or in MV CTR-A (gray bar) evaluated by qRT-PCR. *, p < .05 versus MV CTR-A. Data in (A) and (B) are representative of three different experiments. Data were normalized to GAPDH mRNA and to one for CTR-A and MV CTR-A. (C): Analysis of HepG2 apoptosis by terminal dUTP nickend labeling assay (TUNEL) assay. Cells were incubated with vehicle alone (HepG2), with 10 ng/ml vincristine, or with 30 μg/ml of MV CTR-A, MV DCR−, or control MV-HLSC. Results are expressed as mean ± SD of five different experiments. *, p < .05 versus HepG2; #, p < .05 versus MV CTR-A. (D) Expression of XBP-1, BCL-XL, BCL-2, DHFR, cyclin D1, E2F-2, and Actin was evaluated by Western blot in HepG2 treated with vehicle alone (HepG2) or with 30 μg/ml of MV-HLSC, MV CTR-A, or MV DCR−, and in human hepatocytes treated with vehicle alone (Hep) or with 30 μg/ml of MV-HLSC. Three experiments were conducted with similar results. (E, F): HepG2 were injected subcutaneously in severe combined immunodeficiency (SCID) mice and treated weekly with 100 μg of MV CTR-A or of MV DCR−. Measure of tumor mass (E) and weight (F) was evaluated at sacrifice. Results are expressed as mean ± SD of five mice for experimental group. *, p < .05 versus MV CTR-A. Abbreviations: AGO2, Argonaute 2 transcript; CTR-A, scramble transfected control HLSC; DCR1, Dicer 1; DCR-kd, Dicer knockdown; Hep, hepatocytes; HLSC, human adult liver stem cell; MV-HLSC, HSLC-derived MV; MV CTR-A, MVs from CTR-A transfected HLSC; MV DCR−, MVs from DCR-kd HLSC; Rq, relative quantification.

Effect of specific microRNA (miRNA) inhibitors and of miRNA mimics on in vitro apoptosis and on modulation of target proteins. (A): terminal dUTP nickend labeling assay (TUNEL) assay of HepG2 transfected with specific miRNA inhibitors and with anti-CTR inhibitor and treated with MV-HLSC. Results are expressed as mean ± SD of three different experiments. *, p < .05 versus anti-CTR alone; #, p < .05 versus anti-CTR+MV-HLSC. (B): The antiproliferative effect of MV-HLSC on HepG2 was compared with the effect of miR451, miR223, and miR31 mimics by 5-bromo-2-deoxyuridine incorporation. (C): The proapoptotic effect of miR451 mimic on HepG2 was evaluated by TUNEL assay and compared with that of MV-HLSC and vincristine. Results are expressed as mean ± SD of six different experiments. *, p < .05 versus SCR. (D): Representative micrographs showing the downregulation of MDR1, MIF, and RAB14 targeted by miR451 in HepG2 incubated with MV-HLSC or transfected with mimic 451. Transfection of HepG2 with anti-miR451 (anti-miR451+MV) but not with anti-CTR abrogated the target downregulation induced by MV-HLSC. (E): Representative micrographs showing the downregulation of E2F-2 targeted by miR31 in HepG2 incubated with MV-HLSC or transfected with mimic 31. Transfection of HepG2 with anti-miR31 (anti-miR31+MV) abrogated the E2F-2 downregulation induced by MV-HLSC. Experiments were repeated three times with similar results. Scale bar = 10 μm. Abbreviations: Anti-CTR, negative control miRNA; CTR, control; MV-HLSC, HLSC derived microvesicles; MDR, multidrug resistance protein 1; RAB14, ras-related protein 14; MIF, migration inhibitory factor; SCR, negative control siRNA transfected HepG2.

Effect of specific microRNA inhibitors on MV-induced inhibition of tumor growth and on modulation of target proteins. (A–D): Tumor growth in severe combined immunodeficiency (SCID) mice injected subcutaneously with 3 × 10⁶ HepG2 and treated weekly with vehicle (▪, HepG2) or with 100 μg of MV-HLSC in the presence of anti-miR31 (⋄) or anti-miR451 (□) or anti-CTR (▴). Tumor mass was determined weekly by caliper (A) and by measure of tumor size (B) and weight (C) at sacrifice. (D): Apoptotic cells within the tumors were evaluated by terminal dUTP nickend labeling assay (TUNEL) assay. Results in (B–D) are expressed as mean ± SD of six mice for experimental group. *, p < .05 versus HepG2; ^§, p < .05 versus anti-CTR+MV-HLSC. (E): Representative micrographs showing the in vivo expression of MDR1, MIF, and RAB14 in HepG2, the downregulation of targets by MV-HLSC (anti-CTR+MV), and the abrogation of downregulation by anti-miR451 treatment (anti-miR451+MV). Treatment with anti-miR451 without MV administration (anti-miR451) did not interfere with protein expression by HepG2. (F): Representative micrographs showing the in vivo expression of E2F-2 in HepG2, its downregulation by MV-HLSC treatment in the presence of anti-CTR inhibitor (anti-CTR+MV), and the abrogation of downregulation by anti-miR31 (anti-miR31+MV). Treatment with anti-miR31 without MV administration (anti-miR31) did not interfere with E2F-2 expression by HepG2. Micrographs are representative of six different experiments. Scale bar = 10 μm. Abbreviations: anti-CTR, negative control miRNA; HLSC, human adult liver stem cell; MV-HLSC, HLSC derived MV; MDR, multidrug resistance protein 1; RAB14, ras-related protein 14; MIF, migration inhibitory factor.

In vitro and in vivo effect of human adult liver stem cell-derived microvesicle (MV-HLSC) on HCC hepatocarcinoma, SupT1 lymphoblastoma, and DBTRG glioblastoma cells. (A): Tumor growth in severe combined immunodeficiency (SCID) mice injected subcutaneously with 3 × 10⁶ HCC (▪ green), SupT1 (▪ red), or DBTRG (▪ blue) and treated weekly with vehicle or 100 μg of MV-HLSC (HCC = ▴ green, SupT1 = ▴ red, and DBTRG = ▴ blue). Measure of tumor mass (B) and tumor weight (C) excised from SCID mice treated with vehicle alone (black bars) or with MV-HLSC (white bars) at sacrifice. Results are expressed as mean ± SD of four mice for experimental group. *, p < .05 versus vehicle alone. Effect of MV-HLSC (30 μg/ml) on in vitro proliferation evaluated by BrdU incorporation (D) and on apoptosis evaluated by TUNEL (E) of HepG2, SupT1, and DBTRG. Vincristine was used as positive control of apoptosis (gray bar). Results in (D) and (E) are expressed as mean ± SD of six individual experiments. *, p < .05 versus vehicle alone. Abbreviations: BrdU, 5-bromo-2-deoxyuridine; HCC, hepatocellular carcinoma cell.