Inhibition of porcine reproductive and respiratory syndrome virus infection by recombinant adenovirus- and/or exosome-delivered the artificial microRNAs targeting sialoadhesin and CD163 receptors

Computational prediction represents an effective strategy for identification of the candidate siRNAs or miRNAs that can be validated experimentally. We analyzed the first 1751-nt sequence of porcine Sn mRNA and the 1511-nt sequence of porcine CD163 mRNA for candidate miRNAs. Ten top-scoring miRNA sequences for each target were reported and 18 of them (Table 1) were selected for amiRNA vector construction (Figure 1A). To facilitate the candidate miRNA validation, the first 1751-bp sequence of the Sn cDNA or the 1511-bp sequence of the CD163 cDNA was cloned in frame with the green fluorescent protein (GFP) coding sequence in pEGFP-N1 vector to produce the reporter vector pSn-GFP or pCD163-GFP (Figure 1B). NIH 3 T3 cells were co-transfected with each amiRNA expression vector and the reporter vector, and the cell culture was assayed for GFP⁺ cells by flow cytometry. By using pSn-GFP-transfected cells as the reference, transfection with each amiRSn expression vector plus the reporter vector resulted in GFP⁺ cell number reductions ranging from 39.6% to 96.3% (Figure 1C). Similarly, transfection with each amiRCD163 expression vector plus the reporter vector led to GFP⁺ cell number reductions ranging from 53.3% to 88.5% (Figure 1D). The most effective amiRSn-2 and amiRCD163-2, as well as an irrelevant amiRcon (Table 1), were selected for further study.

We subcloned amiRSn-2, amiR163-2 or amiRcon expression cassette into Ad vector pShuttle-IRES-hrGFP, and three rAds, namely rAd-amiRSn, rAd-amiR163 and rAd-amiRcon, were generated by transfecting AAV-293 cells with the rAd vectors (Figure 2A). Primary PAMs and PK-15 cells were transduced with different doses of rAds, and the cell cultures were analyzed for GFP⁺ cells since the GFP reporter gene was included in the rAd vectors. Compared to 100% transduction efficiency for PK-15 cells at MOI 100 (Figure 2B), only 62% transduction efficiency was achieved for PAMs at MOI 800 (Figure 2C), which was not increased further by using higher MOI. The different rAd transduction efficiencies for the two pig cell types were confirmed by fluorescent microscopy (Figure 2D). Therefore, the further experiments were carried out at MOI 800 for PAM transduction and MOI 100 for PK-15 cell transduction.

To investigate whether the amiRNAs were expressed in rAd-transduced cells, primary PAMs or PK-15 cells were transduced with rAd-amiRSn, rAd-amiRCD163 or rAd-amiRcon, and the total RNA was extracted 48 h after transduction. At the same time, the exosomes were purified from cell culture medium and the total RNA was extracted for sequence-specific amiRNA detection. The expected 80-nt amiRNAs were detected in all of the three rAd-transduced cells (Figure 3A) and their exosomes (Figure 3B), but not in the mock-transduced cells and their exosomes.

The sufficient knock-down of Sn and CD163 receptors is the precondition for the objective evaluation of anti-PRRSV effects of the receptor-targeted amiRNAs. We compared the knock-down efficiencies of the two viral receptors using three different strategies: rAd transduction, exosome treatment and rAd transduction plus exosome treatment. First, primary PAMs were transduced with rAd-amiRSn, rAd-amiR163 or rAd-amiRcon. After incubation for 48 h, the total RNA was extracted for mRNA detection by quantitative RT-PCR and the cell culture was analyzed for Sn⁺ or CD163⁺ cell number by flow cytometry. Compared to that in rAd-amiRcon-transduced cell cultures, Sn or CD163 receptor mRNA expression in the rAd-amiRSn- or rAd-amiRCD163-transduced cells was decreased by 0.51 or 0.54 fold (Figure 4A), while Sn⁺ and CD163⁺ cell numbers were decreased from 48.7% to 22.7% and from 49.9% to 27.6%, respectively (Figure 4B). Next, primary PAMs were incubated with the exosomes (0.5 mg protein/ml) derived from rAd-amiRSn- or rAd-amiRCD163-transduced PK-15 cells. At 48 h after incubation, the total RNA was extracted for quantitative RT-PCR and cell cultures were harvested for flow cytometry analysis as described. Compared to that in the two control groups, Sn or CD163 mRNA expression in the amiRNA-containing exosome-incubated cells was decreased by 0.41 or 0.46 fold (Figure 4A), while Sn⁺ and CD163⁺ cell numbers were decreased to 33.4% and 36.5%, respectively (Figure 4B). Finally, primary PAMs were transduced first with each rAd, incubated for 48 h with each amiRNA-containing exosomes, the total RNA was extracted for quantitative RT-PCR and the cell cultures were harvested for flow cytometry analysis as described. Compared to that in the two control groups, Sn or CD163 mRNA expression was decreased by 0.73 or 0.73 fold (Figure 4A), while Sn⁺ and CD163⁺ cell numbers were decreased to 18.9% and 19.1%, respectively (Figure 4B).

We used two different quantitative assays to evaluate the additive anti-PRRSV effect between Sn and CD163 receptor-targeted amiRNAs: quantitative RT-PCR and viral titration. For the quantitative RT-PCR, primary PAMs were transduced with rAd-amiRcon, rAd-amiRSn and/or rAd-amiRCD163 as described, and infected with PRRSV strain VR2332 (MOI 0.2) 48 h after transduction. At 24 h post infection, the total RNA was extracted for quantitative RT-PCR using PRRSV ORF7-specific primers (Table 2). Compared to that (5.4 log10) in rAd-amiRcon-transduced cells, the ORF7 copy number in rAd-amiRSn and/or rAd-amiRCD163 transduced cells was decreased by 3.5, 2.3 or 2.6 log10 (Figure 5). Next, primary PAMs were incubated with the exosomes derived from rAd-amiRSn- and/or rAd-amiRCD163-transduced PK-15 cells, infected with PRRSV and the total RNA was extracted for quantitative RT-PCR as described. Compared to that (5.4 log10) in the control group, the ORF7 copy number in amiRSn- and/or amiRCD163-containing exosome-incubated cells was decreased by 3.3, 2.0 or 2.2 log10 (Figure 5). Finally, primary PAMs were transduced with different rAds, incubated with different amiRNA-containing exosomes, infected with PRRSV and total RNA was extracted for RT-PCR as described. Compared to that (5.4 log10) in the control group, the ORF7 copy number in double rAd-transduced and double amiRNA-containing exosome-incubated cells was decreased by 4.2 log10, while the ORF7 copy number in single rAd-transduced and single amiRNA-containing exosome-incubated cells was decreased by 3.3 or 3.6 log10 (Figure 5).

For the viral titration assay, primary PAMs were transduced with different rAds and infected with PRRSV strain VR2332 as described. At different time points post infection, the cells were harvested for PRRSV titration on MARC-145 cells. Compared to that (4.3 log10 TCID₅₀) in the mock- or rAd-amiRcon-transduced cells, the PRRSV titer in rAd-amiRSn- and/or rAd-amiRCD163-transduced cells was decreased by 1.5, 0.9 or 1.0 log10 at 24 h post infection (Figure 6A). Next, primary PAMs were incubated with amiRSn- and/or amiRCD163-containing exosomes, infected with PRRSV and harvested for PRRSV titration as described. Compared to that (4.3 or 4.2 log10 TCID₅₀) in the mock- or amiRcon-containing exosome-treated cells, PRRSV titer in amiRSn- and/or amiRCD163-containing exosome-incubated cells was decreased by 1.2, 0.6 or 1.0 log10 (Figure 6B). Then, primary PAMs were transduced with different rAds, incubated with different amiRNA-containing exosomes, infected with PRRSV and harvested for PRRSV titration as described. Compared to that (4.3 or 4.4 log10 TCID₅₀) in the two control groups, the PRRSV titer in double rAd-transduced and double amiRNA-containing exosome-incubated cells was decreased by 2.0 log10, while the viral titer in single rAd-transduced and single amiRNA-containing exosome-incubated cells was decreased by 1.1 log10 (Figure 6C). The similar additive anti-PRRSV effect between the two amiRNAs lasted for at least 96 h (Figure 6A,B or C). Finally, primary PAMs were transduced with different rAds, incubated with different amiRNA-containing exosomes and infected with PRRSV strain JX-A1, CH-1R or VR2332 as described. At 72 h post infection, the infected cells were harvested for PRRSV titration. Compared to that (4.9-6.2 log10 TCID₅₀) in the two control groups, the PRRSV titers in three viral strain-infected cells were decreased by 2.7, 2.1 and 2.6 log10, respectively (Figure 6D).

Mouse fibroblast cell line L-929 (ATCC CCL-1) was grown in Dulbecco’s Modified Eagle’s Medium (DMEM)/Nutrient Mixture F-12 (Gibco, USA) supplemented with 10% fetal bovine serum (FBS) and 1% non-essential amino acids. Pig kidney cell line PK-15 (ATCC CCL-33), mouse embryo fibroblast cell line NIH 3 T3 (ATCC CRL-1658), AAV-293 cells (Stratagene, USA) and African green monkey kidney cell line MARC-145 cells (ATCC CRL-12231) were grown in DMEM supplemented with 10% FBS and 1% non-essential amino acids. Primary PAMs were prepared from 6-week-old Lancrace pigs as described previously [7] and grown in DMEM/F-12 supplemented with 10% FBS, 1% non-essential amino acids and 20% L-929-conditioned medium. PRRSV strain VR-2332 (ATCC VR2332) is a prototype strain of North American genotype [28]. PRRSV strain CH-1R is an attenuated vaccine strain derived from the traditional Chinese strain CH-1a [29]. PRRSV strain JX-A1 is a highly pathogenic Chinese strain [30]. The three PRRSV strains were propagated and titrated on MARC-145 cells.

The first 1751-nt sequence of porcine Sn mRNA (GenBank: AF505985) and the 1511-nt sequence of porcine CD163 mRNA (GenBank: NM_213976) were predicted using the web-based BLOCK-iT™ RNAi designer (Invitrogen, USA). Among 10 top-scoring sequences reported for each target, 18 of them were selected for miR RNAi design according to their higher knock-down probabilities (star rankings). The top- and bottom-strand oligonucleotides (Table 1) were synthesized, each pair of them (5 μl, 200 μM) was denatured at 94°C for 5 min and slowly annealed at room temperature to form double-strand oligonucleotide.

For construction of the amiRNA expression vectors, each double-strand oligonucleotide was cloned at the Bbs I site of amiRNA expression vector pcDNA-miR [31], in which the amiRNA expression cassette consists of the 5′- and 3′-miR flanking sequences of mouse BIC non-coding mRNA, followed by the poly(A) signal of human herpesvirus TK gene (Figure 1A). The recombinant vectors, namely pcDNA-miRSn (1–9), pcDNA-miRCD163 (1–9) or control pcDNA-miRcon, were used for amiRNA validation.

For construction of the reporter vectors, the cellular RNA was extracted from primary PAMs using RNAiso Plus (KaTaRa, Dalian, China) according to the manufacturer’s instructions. The reverse transcription (20 μl) was performed using RevertAid™ Reverse Transcriptase (Fermentas, USA) by following the manufacturer’s protocol. The PCR (50 μl) was performed using 5 μl of RT product and LA Taq DNA Polymerase (KaTaRa, Dalian, China) according to the manufacturer’s manual. The first 1751-bp Sn cDNA segment was amplified using primers Sn For1 and Sn Rev1 (Table 2). The PCR was carried out at 94°C for 5 min; 30 cycles of 94°C for 45 sec, 60°C for 45 sec and 72°C for 2 min; and a final extension at 72°C for 10 min. The first 1511-bp CD163 cDNA segment was amplified using primers CD163 For1 and CD163 Rev1 (Table 2). The PCR was carried out at 94°C for 5 min; 30 cycles of 94°C for 45 sec, 50°C for 45 sec and 72°C for 1.5 min; and a final extension at 72°C for 10 min. Each PCR product was cloned into pMD18-T vector for sequencing and then at the Xho I/Bam HI site of pEGP-N1 vector (Invitrogen, USA) to produce reporter vector pSn-EGFP or pCD163-EGFP (Figure 1B). The expression of two GFP fusions was checked by transfection of NIH 3 T3 cells and fluorescent microscopy.

For construction of the rAd vectors, the amiRSn-2, amiRCD163-2 or amiRcon expression cassette was excised from the amiRNA expression vector by restriction digestion with Aat II and Pvu I, and cloned into pShuttle-IRES-hrGFP (Stratagene, USA) vector after linerization with the same enzymes. The pShuttle-IRES-hrGFP vector is rendered replication defective by deletion of the E1 and E3 genes [32], and by inclusion of humanized Renilla reniformis GFP (hrGFP) sequence after the internal ribosome entry site (IRES) for follow-up purpose. The resultant rAd vector was called pShuttle-amiRSn-IRES-hrGFP, pShuttle-amiRCD163-IRES-hrGFP or pShuttle- amiRcon-IRES-hrGFP (Figure 2A).

NIH 3 T3 cells were seeded in triplicates at 5 × 10⁵/well on 24-well plates and grown to 70% confluent growth. The cells were transfected with pcDNA-amiRSn (1–9), pcDNA-amiRCD163 (1–9) or pcDNA-amiRcon (2 μg/well) plus pSn-EGFP or pCD163-EGFP (0.4 μg/well). The transfection was performed using Lipofectamine™ 2000 (Invitrogen, USA) according to the manufacturer’s manual. At 24 h after transfection, the cells were trypsinized, washed three times with PBS, diluted to 10⁶ cells/ml and analyzed (10⁴ cells) for total fluorescence and cell numbers on BD FACSAia III (BD, USA). The effectiveness of each amiRNA was expressed as the percent inhibition of Sn-GFP or CD163-GFP fusion expression using pSn-GFP- or pCD163-GFP-transfected as the reference.

The three rAds, namely rAd-amiRSn, rAd-amiRCD163 and rAd-amiRcon, were generated by transfecting AAV-293 cells (Stratagene, USA) with the rAd vectors according to the instruction manual for AdEasy™ Adenoviral Vector System (Agilent Technologies, USA). The rAds were amplified on AAV-293 cells, purified using ViraBind™ Adenovirus Miniprep Kit (CELL BIOLABS, USA) by following the manufacturer’s protocol, and titrated on AAV-293 cells as fluorescent formation units (FFU)/ml. For transduction of PK-15 cells, the cells were seeded in triplicates at 5 × 10⁴ cells/well on 24-well plates, and grown to 80% confluent growth. After two time wash with PBS, the cells were transduced (37°C for 2 h) with different doses (MOI) of rAds. The transduction of primary PAMs was carried out as previously described [33]. Briefly, the cells were cultured for 7 days in the growth medium conditioned with 20% of L-929 cell culture medium. The cells were trypsinized, diluted to 10⁵cells/ml with the same medium, mixed with different doses of rAds and centrifuged at 2000 × g for 1 h at 37°C. After wash again with PBS, the cells were grown for additional 48 h in the fresh medium and assayed for GFP-positive cells on BD FACSAia III or by fluorescent microscopy.

Exosome purification was performed as previously described [34]. Briefly, primary PAMs and PK-15 cells were seeded in T75 flasks and grown to 70% confluent growth in the medium supplemented with 10% exosome-depleted FBS. The cells were transduced with each rAd (MOI 800 for primary PAMs or 100 for PK-15 cells) and incubated for additional 48 h. The cell medium was collected and centrifuged at 4°C, 300 g for 10 min, 1000 g for 30 min, 10,000 g for 45 min to remove the cell debris. After filtration through a 0.22-μm filter membrane, the exosomes were precipitated by centrifugation at 120,000 g for 70 min and suspend in PBS for immediate use or stored at −80°C for later use.

The poly(A)-tailed RT-PCR for amiRNA detection was performed using One Step Prime Script miRNA cDNA Synthesis Kit (KaTaRa, Dalian China) by following the manufacturer’s protocol. Briefly, the total RNA was extracted from rAd-transduced cells or purified exosomes for cDNA synthesis. Each amiRNA cDNA was amplified using the sequence-specific forward primer (Table 1) and the general reverse primer in the cDNA synthesis kit. The PCR was carried out at 95°C for 30 sec, 35 cycles of 95°C for 10 sec, 58°C for 30 sec, 72°C for 10 sec, and a final extension at 72°C for 5 min. The PCR products (8 μl) were analyzed by electrophoresis on a 3% agarose gel.

The real-time quantitative RT-PCR for Sn and CD163 receptor mRNA detection was performed as described previously [35]. Briefly, primary PAMs were transduced in triplicates with each rAd and incubated in the medium with or without the exosomes (0.5 mg protein/ml) purified from the rAd-transduced PK-15 cells. At 48 h after incubation, the total RNA was extracted for reverse transcription using RevertAid^TM Reverse Transcriptase (Fermentas, USA). The Sn and CD163 receptor mRNA expression was analyzed by SYBR green-based real-time quantitative PCR using Roche LightCycler Nano system. The mRNA copy number was normalized by comparing to house-keeping GAPDH copy number. The primer sequences are listed in Table 2.

The real-time quantitative RT-PCR for PRRSV ORF7 copy number detection was performed as previously described [36]. Briefly, primary PAMs were transduced in triplicates with different rAds and/or incubated in the growth medium with or without the exosomes derived from rAd-transduced cells as described. At 48 h after incubation, the cells were infected with PRRSV (MOI 0.2), incubated for additional 24 h and total RNA was extracted for reverse transcription and real-time quantitative PCR as described. The standard curve was generated by comparing to PRRSV ORF7-containing plasmid [36]. The primer sequences are listed in Table 2.

The flow cytometry for detection of Sn and CD163 receptor knock-down was performed as previously described [37]. Briefly, primary PAMs were transduced in triplicates with different rAds and/or incubated for 48 h with or without the exosomes derived from rAd-transduced cells. The cells were stained with Sn- or CD163-specific antibody and analyzed by flow cytometry as described.

Primary PAMs were seeded at 2 × 10⁵/ml on 24-well plates and grown overnight in the medium supplemented with 10% exosome-depleted FBS. The cells were mock-transduced in triplicates or transduced with different rAds and/or incubated for 48 h with or without the exosomes (0.5 mg protein/ml) purified from different rAd-transduced PK-15 cells. The cells were infected with PRRSV strain VR-2332, CH-1R or JX-A1 (MOI 0.2) and harvested at different time points for PRRSV titration on MARC-145 cells as previously described [38].