Development of a microRNA delivery system based on bacteriophage MS2 virus-like particles


L. Wang, Beijing Hospital, No. 1 Dahua Road, Beijing, China
Fax: +86 10 65212064
Tel: +86 10 58115053


Recently, microRNA (miRNA)-mediated RNA interference has been developed as a useful tool in gene function analysis and gene therapy. A major obstacle in miRNA-mediated RNAi is cellular delivery, which requires an efficient and flexible delivery system. The self-assembly of the MS2 bacteriophage capsids has been used to develop virus-like particles (VLPs) for RNA and drug delivery. However, MS2 VLP-mediated miRNA delivery has not yet been reported. We therefore used an Escherichia coli expression system to produce the pre-miR 146a contained MS2 VLPs, and then conjugated these particles with HIV-1 Tat47–57 peptide. The conjugated MS2 VLPs effectively transferred the packaged pre-miR146a RNA into various cells and tissues, with 0.92–14.76-fold higher expression of miR-146a in vitro and about two-fold higher expression in vivo, and subsequently suppressed its targeting gene. These findings suggest that MS2 VLPs can be used as a novel vehicle in miRNA delivery systems, and may have applications in gene therapy.


cell-penetrating peptide


fluorescein isothiocyanate


horseradish peroxidase


interleukin-1 receptor-associated kinase 1




peripheral blood mononuclear cell


RNA interference


small interfering RNA


sulfosuccinimidyl 4-(p-maleimidophenyl) butyrate


trans-activator of transcription


virus-like particle


Since the last century, when the ability of small interfering RNAs (siRNAs) to silence target genes was first demonstrated, RNA interference (RNAi) has become a revolutionary strategy for regulating gene expression in various eukaryotic systems [1]. Many classes of RNA, such as antisense RNAs, siRNAs, and vector-based short hairpin RNAs, have been used in RNAi, and microRNAs (miRNAs), which are 20–24-nucleotide noncoding RNAs, are now considered to constitute a new and promising approach in RNAi [2,3]. miRNAs are initially transcribed from intragenic or intergenic regions as long precursors (primary miRNAs) [4,5]. The primary miRNAs are then cleaved, generating the ∼ 70-nucleotide hairpin intermediate pre-miRNAs. Finally, the nuclear pre-miRNAs are exported to the cytoplasm and processed by Dicer into mature miRNAs. Associating with RNA-induced silencing complexes, miRNAs bind to their target mRNAs and result in the degradation of mRNAs and/or translational repression [6,7]. Several reports have documented miRNA-mediated effective gene silencing in cultured cells and mice [8–10].

Owing to the instability and anionic charge of miRNAs, one of the major obstacles to their application in gene silencing is the availability of an effective delivery system [11,12]. A number of approaches to miRNA delivery have been explored, involving the use of viral or nonviral vector systems [11]. However, the possible low transduction efficiency, potential cytotoxicity and integration-induced tumorigenesis are always concerns when plasmid or viral vectors are used in miRNA expression [11]. Therefore, the development of nonviral delivery systems has become a focal point of current research.

Previous studies have reported bacteriophage MS2 as a form of nonviral delivery system [13–15]. MS2 is an icosahedral bacteriophage with a diameter of 27–34 nm. In vitro, the assembly of this bacteriophage can be achieved by interaction between a 19-nucleotide sequence-specific MS2 cistron, called the pac site, and the bacteriophage coat protein [16]. The self-assembled MS2 has been used as virus-like particles (VLPs), allowing the particles to be loaded with RNA or other therapeutic molecules.

However, current studies on the development of intracellular MS2 VLPs as a vehicle are limited to transporting RNAs, siRNAs, or small-molecule drugs. Also, only a few techniques have used a cell-penetrating peptide (CPP)-mediated delivery approach, which might be more effective in transporting, rather than the covalent decoration of the outer capsid surface with ligands for receptor-mediated endocytosis [17,18].

Here, we report a novel miRNA delivery system based on MS2 VLPs and the trans-activator of transcription (Tat)47–57 peptide. Tat47–57 [YGRKKRRQRRR(47–57)] is derived from HIV-1, and has been demonstrated to be one of the most effective CPPs [19]. Its chemical conjugation to MS2 VLPs provides a convenient way to achieve the intracellular transduction and subcellular localization of miRNAs. As a proof of principle, human pre-miR146a RNA was loaded into MS2 VLP–Tat complex, transferred into cells, and then processed into mature miR-146a. Our data show that high levels of MS2 VLP-induced miR-146a could be detected both in vitro and in vivo. Furthermore, this overexpression of miR-146a could effectively suppress the expression of its targeting genes and protein.


Construction of the MS2 VLP expression system

In this study, we chose a single-plasmid coexpression system for the production of VLPs containing pre-miR146a or negative control RNA, and these two VLPs were designated MS2-miR146a and MS2-miRNC, respectively. The purified MS2-miR146a or MS2-miRNC VLPs were analyzed on 0.8% agarose gel. A single band of approximately 1.5 kb was visible, and showed a similar pattern to that of the control MS2 VLPs that we expressed previously (Fig. 1A) [14,15]. In order to confirm that the MS2 VLPs were produced in Escherichia coli BL21(DE3), the purified MS2 VLPs were observed by transmission electron microscopy. The diameter of the MS2 VLPs was ∼ 25 nm (Fig. 1B).

Figure 1.

 Schematic depicting the process used to generate Tat47–57-conjugated MS2 VLPs that package pre-miRNAs. Pre-miRNAs were loaded into MS2 VLPs by two pac sites, and these particles were then conjugated with Tat47–57 by sulfo-SMPB. (A) The MS2-miR146a and MS2-miRNC VLPs were purified by exclusion chromatography, and analyzed on a 0.75% agarose gel, producing bands of between 1000 and 2000 bp in size. The MS2 VLPs previously expressed in our laboratory previously were used as a positive control [15]. Lane 1: MS2-miR146a VLPs. Lane 2: MS2-miRNC VLPs. Lane 3: control MS2 VLPs. Lane M: molecular mass marker. (B) Identification of MS2 VLPs by transmission electron microscopy (×59 000). The diameter of the MS2 particles was approximately 25 nm. (C) The conjugated MS2 VLPs were analyzed by SDS/PAGE. The MS2 VLPs conjugated with Tat47–57 exhibited slower mobility (∼ 15 kDa) than the unmodified MS2 VLPs (∼ 14 kDa). Lane 1: MS2-miR146a VLPs. Lane 2: MS2-miRNC VLPs. Lane 3: unmodified MS2-miR146a VLPs. Lane M: molecular mass marker.

Packaging of the pre-miR146a or control RNA was further confirmed by RT-PCR and sequencing. A 93-bp band was amplified by RT-PCR from MS2-miR146a or MS2-miRNC VLP RNA extract. These results were subsequently verified by sequencing (data not shown), which indicated that MS2 VLPs were expressed successfully and that the pre-miR146a RNA has been correctly packaged in.

Analysis of the conjugation between MS2 VLPs and Tat47–57

To develop a new miRNA delivery system, we conjugated MS2 VLPs with HIV Tat47–57. Conjugation was achieved between the amino group of the MS2 VLP capsids and the cysteine of the Tat47–57 peptide with the help of sulfosuccinimidyl 4-(p-maleimidophenyl) butyrate (sulfo-SMPB). Twenty per cent denaturing PAGE showed that the molecular mass of MS2-miR146a or MS2-miRNC VLPs was approximately 14 kDa, and the Tat47–57-conjugated VLPs exhibited retarded mobility at about 15 kDa (Fig. 1C) [15].

Uptake efficiency of MS2 VLP–Tat conjugates

Tat47–57-mediated MS2 VLP delivery was evaluated by fluorescence detection. HeLa cells, incubated with Tat-conjugated MS2-miR146a or MS2-miRNC VLPs, exhibited a high level of fluorescein isothiocyanate (FITC) fluorescence. In contrast, no signal could be detected in the cells transfected with sulfo-SMPB-conjugated MS2-miR146a or MS2-miRNC VLPs plus free Tat47–57 peptide, demonstrating that uptake of MS2 VLPs did not occur without conjugated Tat47–57 (Fig. 2).

Figure 2.

 Verification of delivery of the Tat47–57-mediated MS2 VLPs by fluorescence microscopy. (A) Untreated HeLa cells. (B) Incubation with unconjugated Tat47–57 and MS2-miR146a VLP–(sulfo-SMPB) conjugate for 24 h. (C) Incubation with Tat47–57-conjugated MS2-miR146a VLPs for 24 h.

MS2 VLP-mediated mature miR-146a expression and its toxicity in cells

To test whether MS2-miR146a VLPs could induce expression of miR-146a in human cells, we transfected different types of cell lines with certain amounts of MS2-miR146a or MS2-miRNC VLPs, and detected miR-146a expression by real-time PCR. As shown in Fig. 3A, low-level expression of endogenous miR-146a was observed in HeLa, HepG2 and Huh-7 cells, and in peripheral blood mononuclear cells (PBMCs). However, after the transfection of MS2-miR146a VLPs, high levels of miR-146a could be detected in all four lines. This increase could not be induced by MS2-miRNC VLPs.

Figure 3.

 MS2 VLP-mediated mature miR-146a expression in vitro and in vivo. (A) MS2-miR146a VLP-induced miR-146a expression in HeLa, HepG2 and Huh-7 cells and in PBMCs. (B) The miR-146a expression induced by various concentrations of MS2-miR146a VLPs (100, 200 and 500 nm) in vitro. (C) MS2-miR146a VLP-induced miR-146a expression at 24, 48, 72, 96, 120, 144 and 168 h in vitro. (D) Cell toxicity test. The cytotoxicity was expressed as cell viability. (E) MS2-miR146a VLP-induced miR-146a expression at 8, 16, 24, 48, 72, 96 and 120 h in mouse plasma. (F) MS2-miR146a VLP-induced miR-146a expression in plasma, liver, lung, spleen, and kidney tissue. nom., normalized.

In order to analyze the dose-dependent response of MS2 VLP-induced miRNA expression, a series of concentrations of MS2 VLPs were used for transfection. A dose-dependent change in miR-146a level was documented over the range of 0.92-fold to 14.76-fold following transfection with 100–500 nm MS2-miR146a VLPs (Fig. 3B). We also evaluated the time-lapse response of MS2 VLP-induced miRNA expression. After transfection with 200 nm MS2-miR146a, PBMCs were harvested at 24, 48, 72, 96, 120, 144 and 168 h post-transfection, and the miR-146a level was analyzed by real-time RT-PCR. The miR-146a concentration was more than five-fold higher than that in the controls, and the expression level was stable during the course of the 120 h observed (P > 0.05, by t-test; Fig. 3C).

A cellular toxicity test by CCK-8 assay showed dose-dependent cellular toxicity in the MS2 VLPs, and the IC50 was > 10 μm (Fig. 3D). Although the exact IC50 value cannot be determined, owing to the limited concentration of our MS2 VLPs, the cell viability under the maximal dose in this experiment (500 nm) should be higher than 90%.

The metabolism and biodistribution of MS2 VLPs in mice

Pharmacokinetic studies were performed in 7-week-old C57BL/6 mice (n = 3 for each time point). Pharmacokinetic parameters were calculated from geometric mean plasma levels per time point. Area under the curve was calculated by use of the trapezoidal rule, and extrapolated to infinity by adding Cz/λz [20]. Key pharmacokinetic parameters are listed in Table 1. Maximum plasma levels were reached at approximately 8 h, and the terminal half-life was up to 43 h (Fig. 3E).

Table 1.   The metabolism and biodistribution of MS2 VLPs in mice. Cmax, maximal concentration after a single injection; tmax, time point of maximal concentration after a single injection; AUC, area under the curve; t1/2, maximal concentration half-time.
TreatmentCmax (fold change)tmax (h)AUCo-inf (fold change)t1/2 (h)
MS2-miR146a VLPs3.818217.0042.88
MS2-miRNC VLPs0.2585.3136.30

We further analyzed the biodistribution of MS2 VLPs in mice. As expected, high levels of miR-146a could be detected in plasma, lung, spleen and kidney after the injection of MS2-miR146a VLPs (2.28-fold, 1.94-fold, 2.07-fold, and 2.23-fold, respectively). Notably, this increase was not observed in liver tissue (P > 0.05, by one-way ANOVA; Fig. 3F).

MS2 VLP-mediated specific gene silencing

We next explored whether MS2-miR146a VLPs could suppress the expression of its specific targeted genes. The activity of firefly luciferase in the dual-luciferase reporter assay system was used to measure the change in expression. Our data showed only slight inhibition in the cells cotransfected with pmirGLO-miR146a plus control MS2-miRNC VLPs or pmirGLO-mismatched miR146a plus MS2-miR146a VLPs, whereas dramatic suppression was found in HeLa cells cotransfected with pmirGLO-miR146a and MS2-miR146a VLPs, verifying the function of the MS2 VLP-induced miRNAs (Fig. 4A).

Figure 4.

 MS2 VLP-mediated specific gene silencing and protein expression suppression. (A) Evaluation of the inhibitory effect of MS2-miR146a VLPs on specific gene translation with a luciferase reporter system. (B) Detection of miR-146a targeting protein, IRAK1, in MS2-miR146a VLP-treated HeLa cells. (C) Evaluation of the suppressed expression of IRAK1 in MS2-miR146a VLP-treated mouse PBMCs.

Suppressive effect of MS2 VLPs on the expression of miRNA targeting protein

MS2-miR146a VLP-induced protein suppression was verified by western blot. According to previous studies, interleukin-1 receptor-associated kinase 1 (IRAK1) is one of the most important targets of miR-146a. Therefore, it was chosen for evaluation of the suppressive effect of MS2-miR146a VLPs [21,22]. It is noteworthy that the MS2-miR146a VLPs showed effective suppression of IRAK1 in vitro and in vivo, whereas no such change could be observed in the MS2-miRNC VLP-treated or control groups (Fig. 4B,C). Thus, this MS2 VLP-based miRNA delivery system can be constructed and applied successfully.


The development of an efficient miRNA delivery system has been a challenge, owing to many limitations, including rapid extracellular degradation and the lack of a reliable delivery system that is capable of transferring miRNAs into the intracellular space without cytotoxicity. Therefore, in order to transfer pre-miRNAs to the correct place, a bacteriophage MS2 VLP-based delivery system, which has several attractive characteristics that are suitable for our purpose, was developed in this study.

First, the self-assembly and packaging of MS2 has been well studied, so it offers an effective and convenient way to package RNAs, DNAs, drugs or imaging agents in bacteriophage capsids, forming different kinds of MS2 VLPs. For example, we were able to package RNA of up to 3034 nucleotides in these VLPs in a previous study [23]. Furthermore, MS2 VLPs not only act as a carrier in the delivery system, but also protect their RNA from interacting with host molecules and from degradation by nucleases. Owing to rapid degradation by nucleases, naked miRNAs are inherently unstable in biofluids. Thus, a variety of carbohydrate and phosphate modifications, such as 2′-O-methyl, phosphorothioates, and locked nucleic acids, have been used to increase the stability of miRNAs [24,25]. Although nucleotide analogs can significantly enhance the half-lives of the synthetic miRNAs, they can also affect the hybridization kinetics, and consequently have an influence on RNAi. However, our previous results showed that pre-miRNAs packaged into MS2 VLPs were protected by the capsid protein from degradation by DNase I and RNase A [14,23], and overexpression of unmodified mature miRNAs was induced after MS2 VLP transduction. Therefore, the MS2 VLPs fulfill the stability requirements for a new miRNA transfection vector.

Second, it has been conclusively demonstrated that Tat peptide can penetrate the cell membrane in vitro and that its cellular uptake mainly depends on an 11-residue cationic peptide sequence of Tat (Tat47–57) [26]. We therefore propose using it as a cooperator to deliver MS2 VLPs into cells. In support of this assumption, fluorescence detection showed that this Tat-mediated non-cell-specific delivery system worked well; the MS2 VLPs were successfully transferred into a series of cells. Moreover, recent studies have suggested that MS2 VLPs can also be transferred into target cells by chemically or covalently conjugating them with some specific CPPs or ligands [27,28]. It would be interesting to explore whether such targeting methods could improve our miRNA delivery system by increasing the efficiency of delivery to targeted cells or by enhancing the effect of RNAi.

In the current study, we showed that this MS2 VLP-based delivery system could induce widespread expression of mature miRNAs over a relatively long period of time. After the transduction of MS2-miR146a VLPs, miR-146a levels over four-fold to five-fold higher than normal could be observed, and this overexpression was sustained for up to 120 h without significant change. This MS2 VLP-based approach had a more effective and durable influence than the transfection of artificial microRNA with a liposome method (3.57-fold overexpression of miR-146a for 72 h). Induced high-level expression of miR-146a has also been demonstrated in vivo, with a 43-h maximal concentration half-time. These observations of miRNA stability are in accordance with previous reports demonstrating that the average half-life of miRNAs is 119 h (∼ 5 days) [29], and that the half-life of miRNA-loaded RNA-induced silencing complexes could be maintained above 140 h (∼ 6 days) following artificial miRNA induction [30]. Interestingly, we noticed that our delivery system does not have a significant impact on liver tissue. The concentration of miR-146a was only 0.23-fold higher than normal after injection. The exact molecular mechanisms are yet to be elucidated, so further investigations are needed.

miRNA interference has become a routine tool for studies of gene function, and alternatives are available to traditional small-molecule therapies. However, miRNA-based gene therapy faces several challenges, including a lack of optimal delivery systems, poor cellular uptake, cytotoxicity, and off-target effects. However, the delivery system described here may address some of these challenges. As compared with the previously reported viral or nonviral delivery systems, this MS2 VLP-based miRNA delivery approach minimizes enzymolysis in biofluids, has relatively lower toxicity, is able to carry a variety of pre-miRNAs into a number of different cell and organ types, and effectively suppresses the expression of its target gene. In particular, unlike most widely used lipid-based delivery systems, MS2 VLPs do not contain cationic lipids, and therefore may not be subject to some weaknesses attributed to charge. For instance, particles based on neutral lipids are less likely to form aggregates in biofluids, be filtered by the liver, adhere to the endothelium, or be taken up by scavenging macrophages [31]. Therefore, our system is highly flexible and well suited for therapeutic research. For example, previous studies have suggested that miR-146a might be a significant brake on autoimmunity, myeloproliferation, and cancer [21,32]. Ablation of miR-146a expression in mice results in several severe immune-related phenotypes that lead to premature death [33]. It is reasonable to hypothesize that the MS2 miR-146a VLPs may compensate for the reduced expression of miR-146a in pathological states, and therefore control the occurrence and development of some immune-associated diseases. To verify this hypothesis and to find more extensive applications of this system, much further work still needs to be done. Despite these limitations, we expect that the MS2 VLP-based miRNA delivery system will be a step forwards in gene research and therapy.

In conclusion, we have succeeded in constructing a novel delivery system that can induce miRNA expression in vitro and in vivo. Our results also confirm the specific RNAi effect of this delivery approach. Thus, the MS2 VLP-based delivery system may represent a novel tool for RNAi and gene therapy in some diseases.

Experimental procedures

Cells and mice

The human uterine cervix cancer line HeLa, the human HCC cell line HepG2 and Huh-7 cells were cultured in DMEM (Gibco, Grand Island, NY, USA) supplemented with 10% fetal bovine serum (Gibco) at 37 °C in 5% CO2. All three cell lines were obtained from ATCC. Human PBMCs were separated from the blood of healthy donors after written permission had been obtained. The isolation was achieved with EZ-SepTM Human Lymphocyte Separation Medium (Dakewei, Beijing, China), following the manufacturer’s instructions, and the PBMCs were then cultured in DMEM supplemented with 10% fetal bovine serum at 37 °C in 5% CO2.

Seven-week-old C57BL/6 mice with a male to female ratio of 1 : 1 (Vital River Laboratories, Beijing, China) were provided with sterilized food and water and housed in a barrier facility under a 12 : 12-h light/dark cycle. The animal experiments were performed in accordance with currently prescribed guidelines and under a protocol approved by Peking University Animal Care and Use Committee.

Plasmid construction

The cDNA encoding MS2 capsid protein was obtained by PCR from plasmid pMS27 (provided by D. S. Peabody, University of New Mexico School of Medicine) with primers MSF and MSR, and inserted in-frame into plasmid pACYCDuet (Novagen, Gibbstown, NJ, USA) to generate the capsid miRNA packaging vector, pMS. The precursor of miR-146a (pre-miR146a) and two C-5 variant pac sites, which were upstream and downstream of pre-miR146a, were designed to be loaded into MS2 VLPs [23]. Oligonucleotide P146, which encodes pre-miR146a and pac sites, was chemically synthesized (Generay, Shanghai, China), and then inserted into plasmid pMS through Xho1 and Kpn1 restriction enzyme sites to construct the recombinant plasmid pMS-miR146. A similar strategy (with oligonucleotide PNC) was used to construct a control plasmid, pMS-miRNC, in this study. The pre-miR146a cDNA, which was inserted into pMS-miR146, was replaced by a random sequence in this control plasmid. Correct plasmid construction was confirmed by sequencing.

Luciferase reporter vector pmirGLO-miR146a and pmirGLO-mismatched miR-146a were generated from pmirGLO Dual-Luciferase miRNA Target Expression Vector (Promega, Madison, WI, USA), following the manufacturer’s instructions. In brief, miR-146a antisense sequence and miR-146a-mismatched antisense sequence were formed through the annealing of two designed oligonucleotides (R146F and R146R, and R146MF and R146MR, respectively), and then inserted into Nhe1/Sal1-digested pmirGLO vector, generating dual-luciferase reporter plasmids as pmirGLO-miR146a and pmirGLO-mismatched miR-146a. All synthesized sequences and details of the primers are listed in Table 2.

Table 2.   PCR primers used in this study. Restriction sites are underlined. The mismatched sequence in R146MF and R146MR is shown in italics.
PrimerSequence (5′- to 3′)
mir146 Reverse primerGTGCAGGGTCCGAGGT

Production and purification of the MS2 VLPs

The recombinant plasmids pMS-miR146 and pMS-miRNC were transformed into competent cells of the E. coli BL21(DE3) strain. MS2 VLPs were expressed, produced and further purified as previously described [23], and designated MS2-miR146a and MS2-miRNC, respectively (Fig. S1). Finally, these two particles were observed by transmission electron microscopy (JEM-1230; JEOL, Tokyo, Japan) at 100 kV and a screen magnification of ×135 000.

Identification of the packaged RNA in MS2 VLPs by RT-PCR

The total RNA of MS2 VLPs was isolated with a QIAamp viral RNA Mini kit (Qiagen, Valencia, CA, USA), according to the manufacturer’s instructions. RT-PCR reactions were performed as described elsewhere [23], with the appropriate primers listed in Table 2. Primer C146R/CNCR was used in the RT reaction, and PCR consisted of 94 °C for 5 min, and 35 cycles of 45 s at 95 °C, 30 s at 56 °C, 30 s at 72 °C, and 10 min at 72 °C, with primers CF and C146R/CNCR for identification of packaged RNA in MS2-miR146a and MS2-miRNC VLPs. PCR products were then purified and ligated with the pEGM-T Easy plasmids (Promega, Madison, WI, USA) for verification by sequencing.

Conjugating the MS2 VLPs with Tat peptide and FITC labeling

The Tat peptide [47YGRKKRRQRRR(47–57)] was generated by solid-phase synthesis (Gima, Shanghai, China). A cysteine was added at the N-terminus of the Tat peptide to provide a free sulfhydryl group [34]. MS2 VLPs (MS2-miR146a and MS2-miRNC) (1.5 mg) were conjugated with sufficient Tat peptide by sulfo-SMPB (Thermo, Rockford, IL, USA), according to the manufacturer’s instructions. In order to assess the efficiency of conjugation, the products were analyzed by 20% denaturing SDS/PAGE. Parts of the MS2-miR146a VLP–(sulfo-SMPB)-Tat and MS2-miR146a VLP–(sulfo-SMPB) conjugates were fluorescently labeled, according to the manufacturer’s instructions, with FITC (Sigma, St Louis, MO, USA). Labeled conjugates were stored at 4 °C in a light-proof container.

Fluorescence assay

HeLa cells were cultivated in six-well cell culture plates at 75% confluency 24 h before transfection, and 200 nm FITC-labeled MS2-miR146a VLP–Tat was added to the medium. The cells were washed three times with NaCl/Pi to remove the MS2-miR146a VLP–Tat-containing medium after 6 h of incubation. Uptake and intracellular distribution of the complex were monitored with a fluorescence microscope with appropriate filters. Unconjugated Tat47–57 and MS2-miR146a VLP–(sulfo-SMPB) were used as negative controls.

Quantitative real-time PCR

Cells (5 × 105) were cultured in each well of the six-well plates, and unlabeled MS2-miR146a or MS2-miRNC VLPs were then added to each well with: (a) various concentrations of MS2 VLPs (100, 200 or 500 nm) for 24 h in HeLa cells; (b) 200 nm MS2 VLPs in various cell lines (HeLa, HepG2, Huh-7, or PBMSc) for 24 h; or (c) 200 nm MS2 VLPs for various periods of time (24, 48, 72, 96 or 120 h) in PBMCs. In order to demonstrate the validity of MS2-miR146a VLP, 10 nm miR-146a mimic (Qiagen) was transfected with HiperFect Transfection Reagent (Qiagen) into PBMCs at the same time. Total RNA was extracted from harvested cells with Trizol (Invitrogen, Carlsbad, CA, USA), according to the manufacturer’s protocol. For miRNA detection, stem–loop RT primers were designed on the based of these sequences as previously described (Table 2) [35,36]. The miR-146a expression levels were quantified by real-time PCR with the PrimeScript RT reagent Kit (Takara, Otsu, Shiga, Japan) and SYBR Premix Ex TaqII Kit (Takara). Briefly, approximately 50 ng of small RNA from each sample was reverse-transcribed to cDNA with stem–loop RT primer and U6 reverse primer. Real-time PCR was then performed on a StepOnePlus Real-Time PCR System (Applied Biosystems, Foster City, CA, USA). All reactions were run in duplicate. In this study, U6 RNA was chosen as an miRNA internal control. The relative expression levels of miRNAs were calculated with the 2−ΔCt method [37], and the differences in miRNA concentrations between treated and control groups were expressed as fold changes.

Cell toxicity test

HeLa cells (104) were inoculated onto a 96-well plate with various concentrations of MS2-miR146a VLPs (from 1 nm to 10 μm) and incubated overnight. After the seeded cells had been rinsed with NaCl/Pi, 10 μL of CCK-8 solution (Dojindo, Mashikimachi, Kumamoto, Japan) was carefully added to each well. The 96-well plate was incubated for 2 h at 37 °C. The absorbance of each well plate was measured at a wavelength of 450 nm with an optical density reader, and each test was performed in triplicate.

MS2 VLP metabolism and biodistribution in vivo

MS2-miR146a or MS2-miRNC VLPs were diluted in sterile water and injected into the tail vein of C57BL/6 mice. Injections contained 100 μg of Tat-conjugated MS2-miR146a or MS2-miRNC VLPs in a volume of 50 μL. Blood was collected at multiple times postinjection, and plasma was separated and kept at −20°C. Also, organs were collected 24 h after injection, and immediately frozen in liquid nitrogen for storage at −80 °C. Because of the high sequence homology of control U6 RNA between human and mice, miR-146a expression levels in mouse serum and organs were quantified by quantitative real-time PCR as described above. C57BL/6 mice treated with placebo (50 μL of physiological saline) were taken as negative controls, and all reactions were run in duplicate.

Firefly luciferase activity assay

Firefly luciferase activity was assessed with a Dual-Luciferase Reporter Assay System (Promega). HeLa cells were first transfected with pmirGLO-miR146a or pmirGLO-mismatched miR-146a luciferase reporter vector with Lipofectamine 2000 transfection reagent (Invitrogen), incubated with Tat-conjugated MS2 VLPs, and harvested as described above. Then, the intracellular fluorescence was determined according to the manufacturer’s instructions. Each experiment was performed in triplicate, and the luciferase activity was defined as the ratio of the reporter firefly luciferase activity to the internal control Renilla luciferase activity.

Western blot analysis

The MS2 VLP-treated HeLa cells and the PBMCs separated from murine spleen tissue of MS2 VLP-injected mice were harvested and lysed in SDS sample buffer. Cell pellets were then homogenized by brief sonication, and the protein concentration was determined with the bichinchoninic acid method [38]. Forty micrograms of total protein was separated on 20% polyacrylamide gels and transferred onto 0.45-μm poly(vinylidene difluoride) membrane in a buffer containing 25 mm Tris/HCl (pH 8.3), 192 mm glycine, and 20% methanol, and blocked with 5% fat-free dry milk in NaCl/Pi for 1 h. These membranes were incubated with specific primary antibodies and detected with horseradish peroxidase (HRP)-labeled secondary antibodies. The primary antibodies were mouse mAb against IRAK1 (Santa Cruz, Santa Cruz, CA, USA) and rabbit mAb against β-actin (Santa Cruz), which were used at a dilution of 1 : 1000. The secondary antibody were the following mAbs: HRP-linked sheep anti-(mouse IgG) (Univ-bio, Shanghai, China) and HRP-linked sheep anti-(rabbit IgG) (Univ-bio), which were used at a dilution of 1 : 5000. β-Actin protein was chosen as an internal control, and the signal was revealed with the quantity one imaging system (Bio-Rad, Richmond, CA, USA).


The authors thank D. S. Peabody (University of New Mexico School of Medicine) for supplying the MS2 capsid protein gene. We are also grateful to S. Sun (Guang’anmen Hospital) for helpful advice. The authors declare that they have no competing interests.