Regulation of influenza A virus infection by Lnc‐PINK1‐2:5

Abstract Influenza virus causes approximately 291,000 to 646,000 human deaths worldwide annually. It is also a disease of zoonotic importance, affecting animals such as pigs, horses, and birds. Even though vaccination is being used to prevent influenza virus infection, there are limited options available to treat the disease. Long noncoding RNAs (lncRNAs) are RNA molecules with more than 200 nucleotides that do not translate into proteins. They play important roles in the physiological and pathological processes. In this study, we identified a novel transcript, Lnc‐PINK1‐2:5 that was upregulated by influenza virus. This lncRNA was predominantly located in the nucleus and was not affected by type I interferons. Overexpression of Lnc‐PINK1‐2:5 reduced the influenza viral mRNA and protein levels in cells as well as titres in culture media. Knockdown of Lnc‐PINK1‐2:5 using CRISPR interference enhanced the virus replication. Antiviral activity of Lnc‐PINK1‐2:5 was independent of influenza virus strains. RNA sequencing analysis revealed that Lnc‐PINK1‐2:5 upregulated thioredoxin interacting protein (TXNIP) during influenza virus infection. Overexpression of TXNIP reduced influenza virus infection, suggesting that TXNIP is an antiviral gene. Knockdown of TXNIP abolished the Lnc‐PINK1‐2:5‐mediated increase in influenza virus infection. In conclusion, the newly identified Lnc‐PINK1‐2:5 isoform is an anti‐influenza lncRNA acting through the upregulation of TXNIP gene expression.

influenza A virus (IAV), leading to the evolution of antigenically novel viral strains that can efficiently replicate in humans. Being such a disease of high pathogenicity and the ability to spread, it is crucial to understand the regulation of influenza virus infection in order to formulate a control strategy.
Only 2% of the human genome encodes for proteins. However, approximately 70% of them are transcribed into RNAs. Based on the size, the protein-noncoding RNAs (ncRNAs) are classified into small ncRNAs (<200 nt) and long ncRNAs (lncRNAs) (>200 nt). 6 Most lncRNAs are transcribed by RNA polymerase II. They share common features with the mRNA such as 5′-capping, splicing, and polyadenylation. The position of the lncRNAs can be in sense or antisense orientation to its neighbouring protein-coding genes, and within introns or in intergenic regions of the genome. Because of the low binding fidelity property of the RNA polymerase and the low conservation between species, lncRNAs were considered as 'transcriptional noise' earlier. LncRNAs have been recently recognized for their crucial functional importance in physiological and diseased conditions. 7 They have specific expression in different cell types, are localized in different subcellular compartments, and are associated with many diseases. 8 LncRNAs execute different types of biological regulatory mechanisms-as signals: causing translational modulation of mRNAs following sequence-specific recognition; as decoys: targeting of chromatin modifiers to DNA through the formation of RNA-DNA hybrids; as a guide: targeting and sequestration of host factors through RNA secondary structures; and as scaffolds: bringing multiple proteins together in order to form functional ribonucleoprotein complexes. 9 We have previously reported dys-regulated lncRNAs in influenza virus-infected human lung epithelial cells via RNA sequencing analysis and found that PSMB8-AS1 is a proviral lncRNA that is induced by IFNβ1. 10 In this study, we identified the new transcript, named as Lnc-PINK1-2:5 that is the major transcript in lung epithelial cells.

| Cell culture
Human alveolar epithelial A549 cell line, HEK cells containing   as previously described. 10 The viral stocks were aliquoted and stored in screw cap tubes at −80°C. The titres of the viral stocks were measured by plaque assay.

| Virus titre determination
MDCK cells were seeded in 6-well plates at a density of 5 x 10 5 cells per well. The next day, cells were washed with sterile Dulbecco's phosphate buffered saline without calcium and magnesium (DPBS) (Lonza, Walkersville, MD, USA) twice. A series of ten-fold dilutions of virus stock ranging from 10 −3 to 10 −8 were prepared in serum-free medium with 1 μg/ml L-1-tosylamide-2-phenylethyl chloromethyl ketone-treated trypsin (TPCK-trypsin). Cells were incubated with 800 µl of each diluted virus stock for 1 h at 37°C in a 5% CO 2 incubator and then overlaid with 2x DMEM and heated 2% sea plaque

| RNA sequencing
RNAs were isolated from 3 vector control-and 3 Lnc-PINK1-2:5overexpressing A549 cells infected with PR/8 at an MOI of 0.01 for 48 h. RNA sequencing (RNA_seq) was performed as previously described. 10 TopHat2 was used to directionally map the paired end reads to the genomic loci of lncRNA (GRCh38 /hg18). Dysregulated mRNAs were identified using cuffdiff analysis based on a fold change of ≥2 and a false discovery rate of <0.05.

| Interferon and Poly(I:C) treatment
A549 cells were seeded in 12-well plates and cultured in F12K medium with 10% FBS and 1% PS overnight. The cells were treated with Poly(I:C) at 100 ng/ml for 24 h or human IFNβ1α (#11415-1, PBL Assay Science, Piscataway, NJ, USA) at various doses for different times. RNA was extracted for real-time PCR analysis.

| Isolation of cytoplasmic and nuclear RNAs
Cytoplasmic and nuclear RNAs were prepared using a Cytoplasmic and Nuclear RNA Purification Kit (Catalog #21000, Norgen Biotek Corporation, Thorold, ON, Canada) from A549 cells. cDNA was prepared using 500 ng RNA and real-time PCR was performed. β-actin (ACTB) and glyceraldehyde 3-phosphate dehydrogenase (GAPDH) were used as cytoplasmic markers and U2 small nuclear RNA (U2 snRNA) was used as a nuclear marker.
The amounts of Lnc-PINK1-2 transcripts along with cytoplasmic and nuclear markers in individual fractions were calculated using the formula 2 −ct and then multiplied by a dilution factor (total ng of RNA in each fraction/500). The distribution percentage of each gene was calculated as target gene amount in individual fraction divided by the sum of target gene amounts in cytoplasm and nuclei × 100.

| Real-time PCR
One µg of the total RNA was reverse-transcribed into cDNA using The comparative ΔCt method using the equation 2 −ΔCt was used to calculate the relative gene expression levels.

| Absolute real-time PCR quantification
cDNA synthesis using one µg of the total RNA was done as described above. A conventional PCR using GoTaq ® DNA Polymerase was performed. The PCR products were purified according to QIAGEN MinElute Gel extraction Kit (Catalog #, 28606 Germantown, MD, USA) and DNA concentrations were measured. Copies/ml of GAPDH and target genes were calculated using the formula; (6.023 × 10 23 × 10 −6 × concentration of the purified PCR products (ng/µl))/molecular weight of the PCR product. PCR was performed at the standard curve mode using the serially diluted known template cDNAs (10 0 to 10 8 ) and unknown samples. Absolute copies of target genes were normalized to GAPDH.

| Western blot
Protein concentration from cell lysates were quantified using Bio-

| Immunofluorescence staining
Primary HSAECs were seeded in SAGM complete medium at a were present in both cytoplasm and nuclei, whereas Lnc-PINK1-2 isoform 3, 4, and 5 were mainly localized in nuclei ( Figure 1E).
Since Lnc-PINK1-2:5 is the major transcript in the lung epithelial cells, we chose it for further investigation. By examining strand-specific genome browser screen shot of the locus from the RNA-seq data, we can see a clear separation of lnc-PINK1-2:5 from the opposite strand of the coding HP1BP3 gene (Figure 2A).

| Lnc-PINK1-2:5 reduces influenza virus replication by upregulating TXNIP
In order to understand the mechanisms of lnc-PINK1-2:5 action, we performed RNA-seq analysis to identify the genes changed in

| DISCUSS ION
LncRNAs have been increasingly recognized to play a crucial role during influenza infection. 28 In this study, we identified a new transcript, Lnc-PINK1-2:5 that was upregulated by influenza virus. We  Our current studies showed that Lnc-PINK1-2:5 expression was induced by influenza virus, but not IFNβ1α, which is supported by the observations that JAK1/2 and STAT1 inhibitors had no effects on IAV-induced Lnc-PINK1-2:5 expression. Furthermore, NF-κB inhibitor also did not reduce IAV-induced Lnc-PINK1-2:5 expression.
However, we found that c-Myc inhibitor inhibited IAV-induced Lnc-PINK1-2 expression. These results suggest that Lnc-PINK1-2:5 expression is regulated by c-Myc, but not type I IFN signalling.
Most lncRNAs that have been studied so far regulate influenza virus replication through the modulation of host immune responses. 28 LncRNA NRAV promotes IAV infection by negatively  37 There are few lncRNAs that are known to act by the mechanisms independent of immune response modulation. Two lncRNAs IPAN and PAAN have been found to stabilize the RNA polymerase PB1 and PA, respectively, and promote the viral replication. 38,39 In this study, we found that Lnc-PINK1-2:5 is an antiviral lncRNA because overexpression of Lnc-PINK1-2:5 resulted in a decreased viral replication as indicated by the viral RNA, protein and titre levels and knockdown of Lnc-PINK1-2:5 had an opposite effect. We also provide evidence that Lnc-PINK1-2:5 exerts its antiviral activity via the upregulation of TXNIP, which is supported by (1) TXNIP overexpression decreased and TXNIP knockdown increased influenza virus infection, suggesting that TXNIP is an antiviral factor, and (2) knockdown of TXNIP abolished the antiviral activity of Lnc-PINK1-2:5.
TXNIP is a redox protein that regulates the functions of pancreatic β-cells. 40 It is a negative regulator of thioredoxin, which plays an important protective role against oxidative stress. 41 TXNIP is identified as an NLRP3-binding protein, mediating oxidative stress and ER stress-mediated activation of NLRP3 inflammasomes. [22][23][24] In an influenza virus polymerase interaction network study using the yeast two-hybrid system, TXNIP was found to interact with the viral polymerase protein PB2. 42 However, this interaction was not F I G U R E 7 Knockdown of TXNIP abolishes Lnc-PINK1-2:5 effect. A549 cells were transduced with lentivirus expressing TXNIP shRNA or its vector control (VC), and/or Lnc-PINK1-2:5 or its VC at an MOI of 100. When two viruses were involved, MOI of 50 were used for each virus. The cells were then infected with a PR/8 at an MOI of 0.01 for 48 h. TXNIP and viral proteins were determined by western blot and quantified. mRNA and protein levels were normalized to β-actin and expressed as a ratio to condition 1 (blank) or 2 (blank with PR/8). Viral titres were determined by plaque assay. Data was shown as means ± SE. n = 3 independent experiments. *p < 0.05, **p < 0.001, ***p < 0.0001 (One-way ANOVA followed by Tukey's comparison) further validated, and whether this interaction influences the polymerase activity remains to be determined.

ACK N OWLED G M ENTS
We express our thanks to Dr. Gillian Air from the University of

DATA AVA I L A B I L I T Y S TAT E M E N T
The data that support the findings of this study are available in GEO database (GSE17 9747) at NCBI.