Human miR-3145 inhibits influenza A viruses replication by targeting and silencing viral PB1 gene

Prediction of candidate miRNAs targeting influenza viral genes

Three subtypes of influenza A viruses including pH1N1 (A/Thailand/104/2009) [accession no. GQ169381-5, GQ205443, GQ229379 and GQ259597], H5N1 (A/Thailand/NK165/2005) [accession no. DQ372591-8] and H3N2 (A/Thailand/CU-H1817/2010) [accession no. CY074963-70] were used to predict human miRNAs that target them. Two web-based programs, miRBase^8–12 (www.mirbase.org) and RNAHybrid¹³ (http://bibiserv.techfak.uni-bielefeld.de/rnahybrid/) were used to screen the human miRNAs that target viral genomes based on the principles of miRNA-target recognition.¹⁴ Hybridization patterns between the miRNAs and their target mRNAs can be classified into 5′ canonical, 5′ seed, and 3′ compensatory. Therefore, criteria for the selection of miRNAs targeting influenza viral genes is based on effective hybridization patterns (5′ canonical, 5′ seed, or 3′ compensatory) and minimum free energy (MFE) for base pairing less than −17.5 kcal/mol. Details of the computational prediction method was described in previous work.¹⁵

Construction of a reporter vector

Total viral RNA was extracted by the Guanidium-isothiocyanate method and then reverse transcribed by random hexamers with SuperScript® III reverse transcriptase (Invitrogen, Waltham, Massachusetts, USA) following the manufacturer’s instruction. The targeting region fragment on viral PB1 genes was amplified by using NheI/PB1_ F and XhoI/PB1_R primers (Table 1). The thermal profile included an initial denaturation at 95℃ for 3 min, 40 cycles of amplification (95℃ for 15 s, 60℃ for 15 s, and 72℃ for 45 s) and final extension at 72℃ for 7 min. These primers were designed with restriction sites of NheI and XhoI for ligation with a pGL3MS2/Basic reporter vector containing firefly luciferase (FLuc) as a reporter gene. The pGL3MS2/Basic vector and PB1 PCR product were digested by restriction enzymes NheI and XhoI (Thermo Scientific, Waltham, MA, USA). The digested products were analyzed by agarose gel electrophoresis and purified by HiYield™ Gel Extraction kit (RBC Bioscience, New Taipei City, Taiwan). Then, the partial PB1 DNA fragment was ligated into XhoI and NheI sites at 3′-UTR of a firefly luciferase gene within a pGL3MS2/Basic vector by T4 DNA ligase (Thermo Scientific). This constructed vector was called pGL3MS2/Basic_PB1.

Construction of silencing vectors

The pSilencer 3.0-H1 (Ambion®, Foster City, CA, USA) was used as a parent vector to construct the miRNA expression vector and positive silencing vector. For pSilencer_miR-3145 (encodes for hsa-miR-3145 short hairpin RNA), two oligonucleotides (1 µg/µL each) – miR-3145_TS (top strand) and miR-3145_BS (bottom strand) – were mixed in 1X annealing buffer then denatured at 90℃ for 3 min, followed by annealing at 37℃ for 1 h (Table 1). After that, the annealed fragment was ligated into BamHI and HindIII restriction sites within a pSilencer3.0-H1 according to the manufacturer’s protocol. For pSilencer_FLuc (encodes for short hairpin RNA against firefly luciferase reporter gene), two oligonucleotides – shRNA-Luc_TS and shRNA-Luc_BS (Table 1) – were annealed and then inserted into pSilencer3.0-H1 with the same protocol as described above.

Plasmids propagation

Plasmids (pGL3MS2/Basic, pGL3MS2/Basic_PB1, pSilencer_scramble, pSilencer_FLuc and pSilencer_miR-3145) were transformed into competent cells (E. coli strain DH5α) (RBC Bioscience, New Taipei City, Taiwan) by a heat shock protocol and then a colony was selected based on ampicillin-resistant characteristics. Nucleotide sequencing was performed in order to confirm the fidelity of each recombinant vector. The concentration of each plasmid was measured by NanoDrop 1000 spectrophotometer (Thermo Scientific).

Cell culture

Human lung adenocarcinoma epithelial cells (A549) were cultured in DMEM (Thermo Scientific) containing 10% heat-inactivated fetal bovine serum (Thermo Scientific) and 1% (v/v) antibiotic-antimycotic (Gibco, Waltham, Massachusetts, USA) at 37℃ under 5% CO₂ with 95% air atmosphere.

Plasmid transfection for 3′-UTR reporter assay

In 3′UTR reporter assay, A549 cells were seeded at 8 × 10³ cells/well in medium without antibiotic-antimycotic in 96 well-plates and incubated for 24 hours. For transfection into each well, three plasmids including 100 ng of a reporter vector (pGL3MS2/Basic_PB1 or pGL3MS2/Basic), 50 ng of a silencing vector (pSilencer_scramble or pSilencer_miR-3145 or pSilencer_FLuc) and 10 ng of a control vector (pCMV_RLuc) were diluted with Opti-MEM (Gibco, Waltham, MA, USA) and then co-transfected into the A549 cells by using 0.4 µL of Lipofectamine®2000 (Invitrogen, Waltham, MA, USA) according to the manufacturer’s protocol. The transfected cells were incubated in a humidified atmosphere with 5% CO₂ at 37℃ for 72 hours and then harvested. The Renilla luciferase activity obtained from pCMV_RLuc was used for normalization of the relative luciferase assay.

Dual Luciferase Assay

The dual luciferase assay was performed by using Dual-Luciferase® Reporter Assay System (Promega, Madison, WI, USA) according to the manufacturer’s instruction. Briefly, cells were washed with 100 µL of phosphate buffer saline and then lysed by adding 20 µL of the Passive Lysis Buffer. The suspensions were then transferred into a white opaque 96-well plate (Costar®). After that, 100 µL of Luciferase assay reagent II (LARII) was added into each well. The emissions of firefly luciferase activity at 560 nm were measured by Synergy HT Microplate reader (BioTek, Winooski, VT, USA). After that, 100 µL of Stop and Glow reagent was added in order to stop firefly luciferase and then measure Renilla luciferase activity at 480 nm. The assay was performed in triplicates in at least three independent experiments. The relative luciferase activity was calculated using signal intensities of firefly luciferase from a reporter vector divided by Renilla luciferase from a control vector.

Plasmid transfection for silencing of viral genes

In a 24-well plate, A549 cells were seeded at 5 × 10⁴ cells/well in DMEM medium without antibiotic-antimycotic. After 24 h of incubation, 600 ng of a silencing vector (pSlilencer_ scramble or pSlilencer_miR-3145) was diluted with Opti-MEM (Gibco, Waltham, Massachusetts, USA) and then individually transfected into the A549 cells by using 1.25 µL of Lipofectamine®2000 (Invitrogen, Waltham, Massachusetts, USA). Then influenza A virus subtypes pH1N1 (A/Thailand/104/2009), H3N2 (A/Thailand/CU-H187/2010) and H5N1 (A/Thailand/NK165/2005) were prepared by diluting the virus stocks with DMEM medium to yield the desired viral titer. After 24 h post-transfection, cell culture media were removed from each well and cells were washed with 1 mL of phosphate buffer saline. After that, 500 µL of each virus suspension (MOI = 0.1) was added into each well and then incubated in a humidified atmosphere with 5% CO₂ at 37℃ for 1 h with occasional shaking every 15 min to allow the virus to adsorb into the cells. After adsorption, the virus suspension was removed from each well and then washed with 1 mL of phosphate buffered saline. Finally, 1 mL of complete DMEM medium was added into each well and cells were incubated in a humidified atmosphere with 5% CO₂ at 37℃ for 48 h. After that, supernatant and cell lysate were collected for further analysis. The experiments were performed in triplicates under a biosafety level 3 (BSL3) laboratory, Faculty of Medicine, Chulalongkorn University. The protocol of this study was approved by Institutional Review Board (IRB No. 121/55), Faculty of Medicine, Chulalongkorn University.

Real-time RT PCR

Total RNA was extracted from infected cell lysates by RBC Real genomics total RNA extraction kit (RBC Bioscience, New Taipei City, Taiwan) and then reverse transcribed with random hexamers by SuperScript® III reverse transcriptase (Invitrogen, Waltham, Massachusetts, USA). The viral PB1 gene and internal control GAPDH gene were quantified by Maxima SYBR Green/ROX qPCR Master Mix (Thermo Scientific). The primers used for amplifications of GAPDH and PB1 genes are summarized in Table 1. Real-time PCR amplification was carried out in Step One Plus™ Real-time PCR Systems (Applied Biosystems, Foster City, CA, USA). The thermal profile included one 3 min cycle at 95℃ and 45 cycles of amplification (95℃ for 15 s, 60℃ for 15 s, 72℃ for 45 s with fluorescence detection at the end of each cycle). After that, a melting curve analysis was performed by increasing the temperature from 60 to 95℃ with a temperature transition rate of 0.5℃/s while continuously collecting the fluorescent signal. The results were analyzed using StepOne™ Software v.2.2 analysis. The relative quantitation was calculated by comparative C_T method to compare the viral gene silencing between cells transfected with pSlilencer_miR-3145 and pSlilencer_scramble control.

Statistical analysis

Data obtained from relative luciferase activity and relative gene expression between each group was analyzed by Student’s t-test. A P ≤ 0.05 was considered statistically significant.

Prediction of candidate human miRNA targeting multiple subtypes of influenza A viruses

Viral genomes obtained from three subtypes of influenza A viruses including pH1N1 (A/Thailand/104/2009), H5N1 (A/Thailand/NK165/2005) and H3N2 (A/Thailand/CU-H1817/2010) were used to predict human miRNA targets based on the miRNA-target recognition patterns (5′ canonical or 5′ seed or 3′ compensatory) and minimum free energy for base pairing (less than -17.5 kcal/mol) criteria. Computational prediction revealed that there were 76 human miRNAs targeting theninfluenza A viruses used in this study. Seventy miRNAs were found to specifically target each subtype of influenza A virus, 21 of which targeted H1N1 subtype, 27 targeted H5N1 subtype and 22 targeted H3N2 subtype (Figure 1). There were 6 miRNAs that targeted multiple subtypes of influenza A viruses including hsa-miR-216 b (target on NA gene of H5N1 and H3N2 subtypes), hsa-miR-3682 (target on NS gene of pH1N1 and H3N2 subtypes), hsa-miR-4513(target on PA gene of pH1N1 and H3N2 subtypes), hsa-miR-4753 (target on PA gene of pH1N1 subtype and PB1 gene of H5N1 subtype), hsa-miR-5693 (target on PA gene of H5N1 and H3N2 subtypes) and hsa-miR-3145 (target on PB1 gene of pH1N1, H5N1 and H3N2 subtypes). The hybridization pattern and MFE of each predicted candidate miRNA are summarized in Table 2.

Table 2.

Summary of candidate human miRNAs targeting multiple subtypes of influenza A viruses

Human miRNAs	Viral Subtype	Viral gene (position)	Hybridization pattern	MFE (kcal/mol)	Pairing pattern
miR-216 b	H5N1	NA (822–843)		−31.3	5′canonical
	H3N2	NA (872–892)		−30.1	5′canonical
miR-3145	pH1N1	PB1 (1551–1570)		−18.2	5′seed
	H5N1	PB1 (1503–1541)		−20.7	5′canonical
	H3N2	PB1 (1531–1549)		−18.1	5′canonical
miR-3682	pH1N1	NS (1011–1028)		−19.1	5′canonical
	H3N2	NS (580–600)		−22.6	3′compensatory
miR-4513	pH1N1	PA (587–609)		−32.3	5′canonical
	H3N2	PA (560–582)		−32.3	5′canonical
miR-4753	pH1N1	PA (1090–1115)		−20.9	5′canonical
	H5N1	PB1 (541–564)		−20.9	5′canonical
miR-5693	H3N2	PA (1353–1370)		−27.9	5′canonical
	H5N1	PA (1353–1370)		−28.7	5′canonical

Open in a new tab

Human miR-3145 targeting 3 subtypes of influenza A viruses

Interestingly, hsa-miR-3145 was the only candidate human miRNA targeting on PB1 gene of all three subtypes of influenza A viruses. The hybridization patterns between hsa-miR-3145 and its viral PB1 target in different subtypes of influenza A viruses were slightly different (Table 2). For pH1N1, the hybridization was classified as “5′seed” with the pairing energy or minimum free energy (MFE) at −18.2 kcal/mol. With H5N1, hsa-miR-3145 was hybridized with a “5′canonical” pattern with MFE at −20.7 kcal/mol. The PB1 of H3N2 targeted a “5′canonical” pattern with MFE at −18.1 kcal/mol. Moreover, alignment of PB1 genes obtained from various strains of human influenza A viruses revealed that PB1 target sites for the seed region of hsa-miR-3145 were conserved among various strains of the viruses (Figure 2). Implying that hsa-miR-3145 might be able to target PB1 of several strains of influenza A viruses. The silencing of PB1 expression by hsa-miR-3145 might have some effects on the replication of influenza A viruses. Therefore, hsa-miR-3145 was selected for further investigation.

Human miR-3145 targeted viral PB1 gene

To investigate the ability of hsa-miR-3145 to target viral PB1 gene, a 3′-UTR reporter assay was performed. According to the conserved target regions in PB1 genes among three viral subtypes, the pH1N1 virus was selected to be the representative for hsa-miR-3145 targeting validation. The pGL3MS2/Basic_PB1 reporter vector was constructed by adding a viral PB1 target region into 3′UTR of the firefly luciferase (FLuc) reporter gene in pGL3MS2/Basic vector. The pSilencer_scramble vector (a scrambled sequence that is not complemented by any gene of the human) and pSilencer_FLuc vector were constructed as negative silencing control vector and positive silencing control vector, respectively. The pSilencer_miR-3145 vector was constructed as a miR-3145 expression vector. The pCMV_RLuc vector was used for Renilla luciferase (RLuc) expression as a transfection control vector. The basis for using this assay is that the hybridization between miR-3145 and the viral PB1 target that linked with 3′-UTR of the firefly luciferase gene will trigger silencing of firefly luciferase activity compared to a control.

Basically, cells in each group were co-transfected with three plasmids including a reporter vector (pGL3MS2/Basic_PB1 or pGL3MS2/Basic), a silencing vector (pSilencer_scramble or pSilencer_FLuc or pSilencer_miR-3145) and a control vector (pCMV_RLuc). Each group was transfected with the same pCMV_RLuc control vector, but differed with respect to the types of reporter vector and silencing vector. Therefore, the transfected A549 cells were separated into six groups (Figure 3). In group 1, cells were transfected with pGL3MS2/Basic and pSilencer_scramble as the control group without target regions in the reporter vector and negative silencing control. In group 2, cells were transfected with pGL3MS2/Basic and pSilencer_FLuc was used as the positive silencing control due to the shRNA expression from pSilencer_FLuc specially targeting to FLuc reporter gene. As the results shown in Figure 3, the dual luciferase activity of group 2 was significantly lower (P = 0.0132) than those found in group 1, showing the effectiveness of the shRNA expression system. In group 3, pGL3MS2/Basic and pSilencer_miR-3145 were transfected into the cells in order to test whether miR-3145 triggered non-specific complementation with a firefly luciferase gene. Even the transfected cells could express hsa-miR-3145, but there was no target region in the reporter vector. As expected, the result revealed no significant difference (P = 0.3162) of the relative luciferase activity between group 3 and group 1, indicating that miR-3145 could not directly bind to the firefly luciferase gene. For group 4, cells were transfected with pGL3MS2/Basic_PB1 and pSilencer_scramble to test a silencing effect triggered by base complementation between an endogenous cellular miR-3145 and PB1 gene within 3′-UTR of the firefly luciferase gene. Results from group 4 showed no significant reduction (P = 0.3604) of the relative luciferase activity between group 4 and group 1, implying that the endogenous cellular miR-3145 was not expressed or was expressed at a very low level that was not sufficient to silence the reporter gene. In another positive silencing control in group 5, cells were transfected with pGL3MS2/Basic_PB1 and pSilencer_FLuc. The significant reduction of luciferase activity in group 5 compared to those found in group 1 and group 4 (P = 0.0026 and 0.0063 respectively) showed the effectiveness of positive silencing control. There was no significant difference of the relative luciferase activity observed between groups 2 and 5, implying that no additional silencing effect occurred when adding the viral PB1 target region into the 3′-UTR of the reporter gene. In group 6, pGL3MS2/Basic_PB1 and pSilencer_miR-3145 were transfected into the A549 cells. In this group, the A549 cells could overexpress hsa-miR-3145 to hybridize with a viral PB1 target region presented at 3′-UTR of the reporter gene. Interestingly, the relative luciferase activity in group 6 was significantly lower than those found in group 1 and group 4 controls (P = 0.0063 and 0.0115, respectively), indicating that hsa-miR-3145 triggered approximately a 63% reduction of the luciferase activity by targeting the viral PB1 segment within the 3′-UTR of the firefly luciferase reporter gene.

Inhibition of influenza A virus replication triggered by hsa-miR-3145

After 24 h of post-transfection with a silencing vector (pSilencer_scramble or pSilencer_miR-3145), the A549 cells were infected with pH1N1 influenza virus (MOI = 0.1). To investigate the antiviral activity of hsa-miR-3145, cell lysate was collected at 48 h post-infection followed by RNA extraction and cDNA synthesis. Expression level of viral PB1 gene was normalized by the amount of host GAPDH internal control gene, which was quantitated by real-time PCR. The infected cells with an overexpression of hsa-miR-3145 showed a significant reduction of viral PB1 RNA expression (P < 0.05) compared to cells transfected with pSilencer_scramble negative silencing control in all three subtypes of viral infection (Figure 4). For the pH1N1 subtype, the PB1 expression was decreased by approximately 84.62% (P = 0.0024) while that in the H5N1 subtype was decreased for 98.95% (P = 0.0004) and PB1 expression of H3N2 subtype was repressed 60.00% (P = 0.0001). Thus, the overexpression of hsa-miR-3145 seemed to inhibit the viral PB1 expression. The PB1 is one of the viral polymerase complexes (PB2, PB1 and PA) required for viral transcription and replication. Therefore, silencing of the viral PB1 gene led to inhibition of viral replication.