Background
Influenza A virus remains a health menace and while yearly statistics vary, ~36,000 deaths and ~220,000 hospitalizations are attributed to influenza each year (mean value) in the United States [
1]. Moreover, the rapidly evolving nature of this segmented RNA virus leads to the emergence of new, unseen subtypes, which have potential to cause a pandemic. Influenza pandemics have occurred at three times in the last century (1918, 1957 and 1968) and once in the current century (2009). Fortunately, the 2009 novel H1N1 influenza virus proved not as pathogenic as initially expected, resulting in fewer deaths than predicted. However, the high transmissibility saw the novel H1N1 readily sweep the globe and resulted in a high global economic burden [
2]. Findings that the highly pathogenic H5N1 avian influenza is able to evolve increased transmissibility in the ferret model [
3,
4], has again emphasized the real possibility that a mutation or recombination event could result in the emergence of a highly pathogenic and easily transmissible influenza virus that will cause a severe and deadly pandemic. Greater understanding of the molecular mechanisms of influenza replication will facilitate the ability to identify novel antiviral targets and develop effective antiviral therapies.
Influenza is a segmented negative strand RNA virus. Each RNA segment is encapsulated by influenza nucleoprotein (NP) and bound by the viral RNA dependent RNA polymerase (RdRP) to form viral ribonucleoproteins (vRNPs) responsible for RNA transcription. Unlike most negative strand RNA viruses, which transcribe RNA in the cytoplasm, influenza transcribes its mRNA in the nucleus. Transcription is primed with a capped 12–15 nucleotide RNA excised from nascent cellular mRNAs by the RdRP in a process termed “cap-snatching” [
5] and polyadenylated by reiterative copying at a poly U stretch near the 5’ end of the vRNA template [
6]. Two viral transcripts, M and NS, are spliced to generate alternative mRNAs. Virus replication requires the nuclear export of spliced mRNAs (M2 and NS2), intron-containing mRNAs (M1 and NS1), and intron-less mRNAs (HA, NA, NP, PB1, PB2, and PA).
The major structure involved in nuclear trafficking is an assembly of nucleoporins located within the nuclear envelope termed the Nuclear Pore Complex (NPC). Large molecular weight complexes require nuclear export signals to export through the NPC. RNAs are transported by interaction with proteins. Studies of retroviral mRNA nuclear export have led to our understanding of two distinct host mRNA nuclear export pathways, represented by the proteins Crm1 and Nxf1. Complex retroviruses such as HIV encode viral Rev protein, which binds the Rev-responsive element (RRE) found within the introns of intron-containing completely un-spliced and partially spliced viral mRNAs, and assists in the export of these viral mRNAs from the nucleus [
7]. Rev contains the first described nuclear export signal (NES) which interacts with host Crm1 to export Rev, along with the bound viral intron containing mRNA, to the cytoplasm [
8]. Simple retroviruses such as Mason-Pfizer monkey virus (MPMV) encode an RNA structure within introns termed the constitutive transport element (CTE), which was used to identify host
n uclear e
x port
f actor 1 (Nxf1, also called TAP) as a cellular mRNA nuclear export factor [
9]. Over-expression of this RNA element blocks most cellular mRNA nuclear export but not Rev dependent mRNA nuclear export [
10]. Conversely, a mutant nucleoporin, which inhibits Crm1, blocks Rev dependent RNA export but not bulk cellular mRNA export or CTE dependent export [
11]. Therefore, the two host cellular proteins, Crm1 and Nxf1, represent two separate mRNA nuclear export pathways.
Research on the mechanisms of influenza mRNA nuclear export is insufficient and results contradictory. RNAi screening in
drosophila cells identified Nxf1 as an essential host factor for influenza mRNA nuclear export [
12]. Additional studies provide evidence of a role for host Nxf1 in export of some but not all influenza mRNAs [
13,
14]. In contrast, another report concludes that influenza NS1 protein inhibits host Nxf1 nuclear export to block expression of host antiviral mRNAs such as IFN mRNAs [
15]. The latter paper suggests influenza mRNA nuclear export is not Nxf1-mediated, but rather Crm1-mediated. While Crm1 nuclear export is utilized by influenza virus for export of viral ribonucleoproteins (vRNPs) during virion assembly [
16], reports support host Crm1 is not used by any influenza mRNAs for export from the nucleus [
13,
14,
17,
18]. The published studies were performed in kidney cells, either Madin-Darby canine kidney cell line (MDCK), baby hamster kidney cell line (BHK), and/or human embryonic kidney cell line (293 T). Given that influenza virus infects cells of the respiratory tract, human lung adenocarcinoma epithelial cell line (A549) are likely a better model cell line for studies of influenza infection. Therefore, we set out to examine influenza viral mRNA export in human lung adenocarcinoma epithelial cell line (A549).
Here we report our results on the role of Nxf1 and Crm1 in influenza intron-less mRNA nuclear export (HA, NA, NP, PB1, PB2, and PA mRNAs). We utilized both inhibition of Nxf1 or Crm1 and direct immuno purification of Nxf1 along with associated RNAs. We find influenza mRNA nuclear export is Nxf1-mediated with the exception of the influenza RNA dependent RNA polymerase encoding mRNAs; PA, PB1, and PB2. Our results in A549 cells differed from our results and published research obtained in 293 T cells [
13] with respect to the export of influenza NP mRNA. This led us to conclude there is a cell type difference in Nxf1-mediated NP mRNA nuclear export: in human lung adenocarcinoma epithelial cell line (A549) NP mRNA nuclear export is Nxf1-mediated while in human embryonic kidney cell line (293 T) NP mRNA nuclear export is Nxf1 independent. It is important to acknowledge cell type differences if the larger goal is to translate data to application.
Although much research suggests Crm1 is not utilized for influenza mRNA nuclear export [
13,
14,
17,
18], in light of the revelation of a cell type difference, we readdressed the role of Crm1 in influenza mRNA nuclear export in A549 cells. Inhibition of Crm1 did not result in significant inhibition of nuclear export of any influenza mRNAs analyzed. This led us to conclude that the influenza RNA dependent RNA polymerase encoding mRNAs; PA, PB1, and PB2, do not export the nucleus via the two defined mRNA nuclear export pathways represented by Crm1 and Nxf1, but instead use an atypical mRNA nuclear export pathway. Defining this pathway will shed light on alternate cellular mRNA nuclear export pathways and may lead to the discovery of novel antiviral targets.
Conclusion
There are three main conclusions from our research results. The first is that Nxf1-mediated nuclear export is required for optimal influenza virus production in all cell types examined. When Nxf1-mediate nuclear export is inhibited by expression of dominant negative Nxf1 protein there is less nuclear export of select influenza mRNAs and less virus production. These data suggest that Nxf1-mediated nuclear export needs to be functional for optimal influenza replication. Further, select influenza mRNAs were found associated with Nxf1 in the cell, strongly supporting use of Nxf1-mediated nuclear export by these select viral mRNAs.
The second main conclusion is the recognition that model cell type can influence molecular mechanisms of influenza infection. We clearly demonstrate that NP mRNA nuclear export is Nxf1-mediated in A549 cells but Nxf1-independent in 293 T cells. This conclusion is important to properly assess research results for relevance to application. While 293 T cells are routinely used in research because of excellent ability to take up DNA, they are kidney cells and thus may not be the best human model cell line to study the molecular mechanism of the respiratory influenza A virus; A549 lung cells may represent a better model. Our study is a needed reminder that cell type can influence molecular mechanisms and this must be taken into consideration when analyzing, compiling, and comparing data.
The third and in our opinion most important main conclusion is that the polymerase encoding mRNAs, PA, PB1, and likely PB2, appear to use neither defined host mRNA nuclear export pathway in both cell types examined. This is an exciting result as it implies these viral mRNAs utilize an uncommon mRNA nuclear export pathway. While there are other RNA nuclear export factors, such as Exportin T to export tRNAs [
27] and Exportin 5 to export pre-micro-RNAs [
28], these factors have not as yet been implicated in export of mRNAs, viral or host. It may be that influenza has hijacked one of these non-coding RNA export pathways to export viral mRNAs. Further, Nxf1 is a member of a family of
n uclear e
x port
f actors, and it could be that influenza polymerase encoding mRNAs have hijacked a less characterized member of this family. It is also possible that the influenza polymerase encoding mRNAs utilize an as yet unidentified and uncharacterized nuclear export pathway which likely function in atypical mRNA nuclear export pathway(s) and may represent feasible targets for future development of innovative antiviral therapies. Identification of the host factors that participate in the nuclear export of influenza PA, PB1, and PB2 mRNAs is a goal of our future research.
Methods
Cells, virus, plasmids, LMB, and antibodies
293 T human embryonic kidney cell line, A549 human lung adenocarcinoma epithelial cell line, and MDCK Madin-Darby canine kidney cell line were purchased from ATCC American tissue culture collection and maintained at 37°C with 5% CO2 in DMEM with 10% FBS. Influenza A/Udorn/307/1972(H3N2) was generated from 12 plasmids using reverse genetics as described [
29]. The plasmids required for reverse genetics were kindly provided by R. Krug. Plasmids encoding dominant negative DN - Nxf1 (TapA17) was kindly provided by B. Cullen [
19] and plasmid encoding FLAG-Nxf1 was kindly provided by J. Steitz [
20]. Plasmid DNA was purified using QIAGEN maxi or mini prep kit per manufacturers protocol. Leptomycin B (LMB) was purchased from Fisher. α-Tubulin, α-SP1, α-HSP90, α-NXF1, and α-FLAG were purchased from Abcam and used per manufacturers instructions. α-NS1 Udorn was kindly provided by R. Krug. Secondary HRP coupled α-Mouse and α-Rabbit were purchased from Pierce and used per manufacturer’s instructions.
Transfection
293 T or A549 cells were grown to approximately 70% confluency in 100 mm dishes or 6-well plates depending on experiment.
pcDNA plasmids encoding FLAG-NXF1 (10 μg) and FLAG-Vector (9 μg) with eGFP (1 μg), or DN-Nxf1 (10 μg) and CMV (9 μg) with eGFP (1 μg) were transfected into cells (100 mm dish) using Mirus transfection DNA to reagent at a ratio of 1:3 (293 T cells) or 1:2 (A549 cells). eGFP was used to monitor transfection efficiency, which was ~70% in A549 cells and ~90% in 293 T cells. 48 hrs post transfection one set was mock infected and the other infected at MOI of 2.5.
LMB treatment
Cells were allowed to incubate with viral inoculum (MOI 1.4 or 2.8) for 1 hour; after attachment viral inoculum was removed and 10nM LMB was added along with the incubation media in treated samples for the remainder of influenza infection period.
Influenza A infection
Cells were infected with MOI of 1.4, 2.5, or 2.8 as indicated. Cells were first washed with PBS and overlaid with viral inoculum for 1 hour with gentle shaking every 15 minutes to ensure cells did not dry out and virus attachment occurred. After 1 hour virus inoculum was removed and replaced with media containing 2.5% FBS. For experiments where virion or virus production was assessed, cells were washed numerous times prior to addition and sampling of media. Furthermore, a control sample was taken at this point to ensure minimal/no virus left behind from inoculum. For examination of mRNA nuclear export under conditions of Nxf1 or Crm1 inhibition, cells were collected at 3.5 hours post infection. For examination of direct Nxf1-mRNA interaction, cells were collected at 7 hours post infection.
Plaque and HA assays
Serial dilutions of media samples were subject to standard plaque or HA assay. For plaque assay, media from triplicate samples were diluted and used to infect confluent MDCK cells. Titers from triplicate trials were averaged and standard error was obtained by calculating the standard deviation of the sample set divided by the square root of the sample set size, and indicated using error bars. Significance was determined using a two-tailed T-Test conducted in Microsoft Excel, and judging any p value less than .05 as significant. For HA assay two-fold dilutions of media was mixed with chicken red blood cells.
Cellular fractionation and isolation of protein and RNA
At 3.5 hours post infection, cells were pelleted by centrifugation and cellular pellets were washed in 5X volume of cell pellet with Reticulate Standard Buffer (RSB: 10 mM Tris HCl pH7.5, 10 mM KCl, 1.5 mM MgCl2) containing protease and RNase inhibitors. Cells were then re-suspended in RSB at 10X the volume of the cell pellet and incubated on ice for 10 minutes. NP-40 was added to a final concentration of 0.2% to disrupt plasma membranes. Visual inspection of the cells before and after addition of NP-40 ensured burst plasma membranes and intact nuclei. Nuclei were pelleted by centrifugation at 300×g for 8 minutes at 4°C. The cytoplasmic extract was collected and the nuclear pellet was re-suspended in Dignam Buffer C without glycerol (20 mM HEPES pH 7.9, 0.42 M NaCl, 1.5 mM MgCl2, 0.2 mM EDTA) and containing protease and RNase inhibitors, to release nuclear molecules. Both nuclear and cytoplasmic extracts were clarified from debris by high-speed centrifugation for 10 minutes at 4°C. An equal amount of 20 mM HEPES pH 7.9, 0.2 mM EDTA was added to the nuclear extract to reduce the total NaCl and MgCl2 concentrations. This was used as cytoplasmic and nuclear protein extracts.
To isolate RNA, equal volume of Phenol/Chloroform/Isoamyl alcohol (25:24:1) was added to a portion of the cytoplasmic and nuclear fractions. Samples were vortexed 4 times for 10 seconds and placed on ice in between. Samples were centrifuged at 13,000 RPM for 10 minutes at 4°C. Aqueous layer was collected and 0.5 volume NH4OAc (7.5 M) and 2X volume 100% EtOH was added and RNA was allowed to precipitate overnight at -80°C. Samples were centrifuged at 13,000 RPM for 20 minutes at 4°C. Pellet was washed in 75% EtOH and centrifuged at 13,000 RPM for 5 minutes at 4°C. EtOH was removed and pellet was allowed to air dry for 10 minutes and resuspended in 10 mM Tris in DEPC H2O; amount dependent on size of RNA pellet. RNA was quantified using a nanospectrophotometer and absorbance at 260.
For immuno purification experiments, cell pellets were resuspended in 1 mL Sonication Buffer (100 mM Tris HCl pH 7.5, 100 mM NaCl, 2.5 mM MgCl2, 0.5% Triton ×-100) containing protease and RNase inhibitors. Cells were lysed using Fisher Scientific Sonic Dismembrator for 30 pulses at 30%, output 3–4. Sonicated materials were loaded onto a 30% sucrose cushion (30% Sucrose, 10 mM Tris HCl pH 7.5, 100 mM NaCl, 2.5 mM MgCl2) and centrifuged at 4000 RPM for 15 minutes at 4°C to clarify total protein extract.
Immunopurification
A portion of total protein extract was incubated with α-FLAG antibody (Stratagene) (1:50) and protease inhibitors for 1 hour at 4°C. Extracts were then incubated with PA/G sepharose beads washed with sonication buffer and containing protease inhibitors, at 4°C overnight. The samples were then spun for 8 seconds at 13,000 RPM at 4°C. Supernatant was collected. Beads were washed 3 times in 1 mL sonication buffer with RNase inhibitors. With each wash samples were then centrifuged for 8 seconds at 13,000 RPM at 4°C. 1/3 of the immuno purification from last wash was saved for protein analysis, and 2/3 was saved for RNA isolation.
RNA isolation
Samples were first subject to protease degradation by incubation with protease K and equal volume of Phenol/Chloroform/Isoamyl alcohol (25:24:1) was added to the resuspended bead immunopurification sample and total extract and samples processed as previously described above.
SDS-PAGE and Western Blot
Protein extracts were separated by SDS-10% PAGE. Proteins were transferred to nitrocellulose using Fisher semi-dry blot apparatus and probed with primary and HRP-conjugated secondary antibodies as indicated. Pierce ECL reagents were used to detect HRP conjugated secondary antibody. Blots were developed using the Chemi-Hi setting on the ChemiDoc™ XRS (BioRad) system and digital images were obtained using Quantity One software. Digital images were exported as raw data TIFF files and image prepared using Adobe Photoshop to crop photos and adjust exposure, and Adobe Illustrator to add figure text labels.
Reverse transcription – PCR
RNA was first quantified by spectrophotometry and separated on a 1% bleach/1% agarose gel to observe rRNA in total and cytoplasmic RNA preparations. For inhibition experiments, 1 μg cytoplasmic RNA was subject to reverse transcription using Promega AMV reverse transcription system per manufacturer’s protocol with oligo dT as primer. For immuno purification experiments, 150-500 ng RNA was subject to reverse transcription depending on RNA recovery; however in all cases equivalent RNA concentration was used for all samples in the RT step.
End point and semi-quantitative PCR
For analysis of mRNAs associated with Nxf1, 10% cDNA from total samples relative to 100% cDNA from IP was used for PCR (for example 1ul total and 10ul IP). For analysis of mRNA in cells inhibited for Nxf1-mediate nuclear export, equal volume cDNA was subject to gene specific PCR. For analysis of mRNAs associated with Nxf1 samples were taken at cycle 25. For analysis of mRNA in cells inhibited for Nxf1-mediate nuclear export samples were taken at sequential cycles as indicated to confirm analysis within the PCR exponential amplification curve. Samples were observed by ethidium bromide 1% agarose gel electrophoresis.
Primers for end point and semi-quantitative PCR:
NP (forward - CCAGAAGAAGTGTCCTTCCG,
reverse - CGTACTCCTCTGCATTGTCTCC),
PB1 (forward - CCCCTGAATCCATTTGTCAGCCATA,
reverse - ATGAAGGACAAGCTAAATTG),
HA (forward - GCTCTGGAGAACCAACATACAA,
reverse - ACAAGGGTGTTTTTAATTACTAATA),
PB2 (forward - CCACCCAGATAATAAAGCTTCTCCCC,
reverse - GTCAGTAAGTATGCTAGAGTCCCG),
PA (forward - ATGACCAAAGAGTTTTTTGAGAATA,
reverse - GTATGGATAGCAAATAGTAGCATTG).
Quantitative PCR
Equal amounts of cDNA were used in triplicate reactions. Quantitative PCR reactions were run in the Applied Biosystems StepOnePlus Real Time PCR system using SYBR Green Master mix (Applied Biosystems), with ROX as the reference dye.
Primers for gene specific qPCR:
NA (forward TGTGTGCTCAGGGCTTGTTG,
reverse CTCCGTGATTCCCTTTCTCATT)
HA (forward ACTGAAGTCAGGATACAAAGACTGGAT,
reverse CCCCAGCAAAACAACACAAA)
NP (forward GTGTGCAACCTGCATTTTCTGT,
reverse TCTGAGGTTCTTCCCTCCGTATT)
PA (forward GGACAAATGGAACATCAAAGATTAAA
Reverse CAGAAGACTCGGCTTCAATCATG)
PB1 (forward GGGAAAGGATACATGAACGAAAGT
reverse ACTTTAGGTCAATGCTTGCTAGCA)
PB2 (forward AATAAAGCTTCTCCCCTTTGCA
reverse CCCTCACATTCACAGTCAATGAA)
PCR cycle
PCR was performed in a standard 3 cycle PCR with denaturation at 95°C for 30 seconds, annealing temperature of 55°C for 30 seconds, and extension temperature of 72°C for 30 seconds.
Statistical Analysis of qPCR Data
Raw CT values were analyzed in Microsoft Excel using the 2ΔCt(control-treated) formula of 2^CTaverage Control sample/2^ CTaverage Treated sample. Due to the robust host shut off that occurs during influenza infection, we were unable to reliably detect a reference gene but rather normalized to total RNA concentration. RNA OD to calculate concentration was taken in duplicate or triplicate and RNA analyzed using gel electrophoresis prior to reverse transcription to ensure equal rRNA concentration and no RNA degradation. Reverse transcription reactions were aliquot from a master mix to ensure all samples obtained equivalent AMV-RT enzyme. Standard error was obtained by calculating the standard deviation of the sample set divided by the square root of the sample set size, and indicated using error bars. Significance was determined using a two-tailed T-Test conducted in Microsoft Excel, and judging any p value less than .05 as significant.
Open Access This article is licensed under a Creative Commons Attribution 4.0 International License, which permits use, sharing, adaptation, distribution and reproduction in any medium or format, as long as you give appropriate credit to the original author(s) and the source, provide a link to the Creative Commons licence, and indicate if changes were made.
The images or other third party material in this article are included in the article’s Creative Commons licence, unless indicated otherwise in a credit line to the material. If material is not included in the article’s Creative Commons licence and your intended use is not permitted by statutory regulation or exceeds the permitted use, you will need to obtain permission directly from the copyright holder.
Competing interests
The authors declare that they have no competing interests.
Authors’ contributions
SL designed and standardized conditions for all qPCR primer sets, performed dominant negative Nxf1 studies in both 293 T and A549 cells (Figure
2) and RT-qPCR analysis of isolated RNA from HMR (Figure
1D). SL also wrote the initial draft of most of the included research results as his Biology Master’s thesis; this was copied, edited to include additional experiments and discussion, and formatted for Virology Journal by LLN. SB designed and standardized conditions for all semi-qPCR primer sets, performed plaque assays for influenza viral titer after inhibition of Nxf1 via DN-Nxf1 expression (Figure
1A), established conditions for immuno purification of FLAG-Nxf1/RNA complexes (Figure
3C) and FLAG-Nxf1-NS1 interaction (Figure
4). VP performed immuno purification experiments of FLAG-Nxf1/RNA complexes (Figure
3A and B). AM carried out the LMB inhibition experiments with assistance from VP. AM obtained the LMB inhibition results (Figure
5). HMR performed dominant negative Nxf1 studies in A549 cells and RT-semi qPCR (Figure
1B and C). LLN conceived and designed the project, allowing for input from all student authors, oversaw all student researchers, participated in qPCR analysis, prepared all figures in the acceptable file format, and wrote the final draft of the manuscript. All authors read and approved the final manuscript.