The authors have declared that no competing interests exist.
Conceived and designed the experiments: PZ HHW SP. Performed the experiments: PZ ASM. Analyzed the data: PZ SP. Contributed reagents/materials/analysis tools: HHW. Wrote the paper: PZ SP.
The cellular protease TMPRSS2 cleaves and activates the influenza virus hemagglutinin (HA) and TMPRSS2 expression is essential for viral spread and pathogenesis in mice. Moreover, severe acute respiratory syndrome coronavirus (SARS-CoV) and other respiratory viruses are activated by TMPRSS2. However, previous studies on viral activation by TMPRSS2 focused on a 492 amino acids comprising form of the protein (isoform 2) while other TMPRSS2 isoforms, generated upon alternative splicing of the tmprss2 mRNA, have not been characterized. Here, we show that the mRNA encoding a TMPRSS2 isoform with an extended N-terminal cytoplasmic domain (isoform 1) is expressed in lung-derived cell lines and tissues. Moreover, we demonstrate that TMPRSS2 isoform 1 colocalizes with HA and cleaves and activates HA. Finally, we show that isoform 1 activates the SARS-CoV spike protein for cathepsin L-independent entry into target cells. Our results indicate that TMPRSS2 isoform 1 is expressed in viral target cells and might contribute to viral activation in the host.
Respiratory viruses pose a significant threat to human health. In particular, annual influenza epidemics are associated with several hundred thousand deaths every year, and interspersed pandemics may wreck even greater havoc [
The surface proteins of influenza A viruses (FLUAV) and coronaviruses, termed hemagglutinin (HA) and spike (S), respectively, facilitate viral binding to host cells and fusion of the viral envelope with a host cell membrane [
Alternative splicing of the messenger RNAs produced from several TTSP genes has been reported [
Here, we show that mRNA encoding isoform 1 is expressed in certain lung-derived cell lines and tissues and that the protein can activate FLUAV and the S proteins of SARS-CoV and MERS-CoV (SARS-S, MERS-S), suggesting that isoform 1 could promote viral spread in the infected host.
Nucleotide and amino acid sequences were analyzed using the BLAST server at NCBI. The alignment of the amino acid sequences of the N-termini of TMPRSS2 isoform 1 (NP_001128571.1) and isoform 2 (NP_005647.3) was constructed using the Clustal Omega software.
Expression plasmids for H1 and H3 FLUAV HA, SARS-S, MERS-S, ACE2 and DESC1 have been described previously [
The cell lines 293T (ATCC CRL-3216), Calu-3 (ATCC HTB-55), Caco-2 (ATCC HTB-37), EA-hy (ATCC CRL-2922), BEAS-2b (ATCC CRL-9609), NCI-H292 (ATCC CRL-184), NCI-H727 (ATCC CRL-5815), A549 (ATCC CCL-185) and COS-7 (ATCC CRL-1651) were grown in Dulbecco’s modified Eagle’s medium (DMEM, PAN Biotech), while LNCaP (ATCC CRL-1740) and NCI-H358 cells (ATCC CRL-5807) were propagated in RPMI-1640 medium (PAN Biotech), and MDCK cells (ATCC CRL-2936) were grown in minimum essential medium (Gibco). All media were supplemented with 10% fetal bovine serum (Biochrome), 100 U/ml penicillin and streptomycin (PAN Biotech). The cells were cultured in humidified atmosphere containing 5% CO2. All cell lines were obtained from collaborators and were regularly checked for mycoplasma contamination. The FLUAV A/PR/8/34 (H1N1) and A/Panama/2007/99 (H3N2) were propagated in the chorioallantoic cavities of 10-days-old embryonated hen eggs (Valo BioMedia, Germany) for 48 h at 37°C. Thereafter, the eggs were euthanized by an overnight incubation at 4°C and the allantoic fluid was harvested. Before propagation in eggs, the FLUAV A/Panama/2007/99 was recovered from a reverse genetics system [
Total RNA was isolated from human cell lines with RNeasy Mini Kit (Qiagen), as recommended by the manufacturer. The Human MTC™ Panel I (Clonetech) cDNA was used to analyze mRNA expression in human organs. This cDNA was obtained from pooled tissue samples from 1–15 Caucasian donors aged 18–69 and representing both sexes. Using a Cloned AMV First-Strand cDNA synthesis kit (Invitrogen) and random hexamers, the first strand cDNA synthesis was performed from 1 μg of total RNA, previously treated with DNaseI (Roche), according to the manufacturer’s protocol. The subsequent PCR was performed with Taq polymerase (New England Biolabs) using gene-specific primers for tmprss2 transcript variant 1 (forward 5’ GTG AAA GCG GGT GTG AGG A 3’ and reverse 5’CTG TGC GGG ATA GGG GTT TT 3’), tmprss2 transcript variant 2 (forward 5’GCG AGG GGC GGG GAG CGC C 3’ and reverse 5’ GGT AGT ACT GAG CCG GAT GC 3’) and GAPDH (forward 5’ ATG GGG AAG GTG AAG GTC GG 3’ and reverse 5’ ATA CTT CTC ATG GTT CAC AC 3’). All PCRs were run for 40 cycles of 30 sec denaturation at 95°C, 30 sec annealing at 58°C, and 30 sec elongation at 72°C. Amplicons were analyzed by agarose gel electrophoresis.
For analysis of the expression of the TMPRSS2 isoforms, 293T cells were seeded into 6-well plates at a density of 2.8 x105 cells/well, cultivated for 24 h and then transfected with plasmids encoding the proteases equipped with an N-terminal myc tag or transfected with empty plasmid as control. After overnight incubation, the medium was replaced with fresh DMEM, and at 48 h post transfection the cells were washed with phosphate-buffered saline (PBS), resuspended in 100 μl of 2 x sodium dodecyl sulphate (SDS) loading buffer per well and then heated at 95°C for 30 min. Protein samples were separated by SDS-PAGE and blotted onto a nitrocellulose membrane (Hartenstein). TMPRSS2 isoforms were detected using a mouse anti-myc antibody (Biomol) as the primary antibody and a horseradish peroxidase (HRP)-coupled antibody (Dianova) as the secondary antibody. Expression of β-actin, detected with anti-β-actin antibody (Sigma Aldrich), served as a loading control. Bound antibodies were detected using ECL Prime Western blotting detection kit (Amersham), according to the manufacturer’s instructions. Image acquisition was performed with a ChemoCam Imager (Intas). For analysis of TMPRSS2 isoform expression by flow cytometry, 293T cells were transfected with plasmids encoding TMPRSS2 isoforms, as described above. At 48 h post transfection, the cells were detached, incubated with ice-cold ethanol for 10 min and stained with anti-TMPRSS2 antibody (Santa Cruz Biotechnology) diluted in 1% saponin for 30 min. The mouse IgG1 (R&D Systems) served as isotype-matched control. Thereafter, the cells were washed three times with PBS and incubated with an AlexaFluor647-coupled anti-mouse antibody (Dianova) diluted in 1% saponin. After 30 min of incubation with secondary antibody, the cells were washed two times with PBS and then fixed with 2% paraformaldehyde. The staining was analyzed with an LSRII flow cytometer (BD Biosciences) and FCS Express 4 Flow Research Edition software (DeNovo Software).
To determine cleavage of viral glycoproteins mediated by TMPRSS2 isoforms, 293T cells were seeded in 6-well plates at 2.8 x105 cells/well, cultured for 24 h and then cotransfected with 6 μg of plasmid encoding FLUAV HA of the H1 or H3 subtype or SARS-S with a C-terminal V5 tag and 0.1 μg of plasmid encoding the indicated proteases, employing the calcium phosphate transfection protocol. At 48 h post transfection, the cells were harvested in PBS and treated with PBS or 250 μg/ml tosylsulfonyl phenylalanyl chloromethyl ketone (TPCK) trypsin (Sigma Aldrich) for 10 min at 37°C and processed for Western blot analysis as described above. The SARS-S protein with a C-terminal V5 tag was detected by staining with mouse monoclonal antibody reactive against the V5 tag (Invitrogen), followed by incubation with an HRP-coupled anti-mouse secondary antibody (Dianova). The FLUAV HA cleavage was detected by staining with a goat anti-FLUAV polyclonal antibody (Millipore) for H1 subtype or with a rabbit anti-H3 HA serum (Immune Technology) and HRP-coupled anti-goat or anti-rabbit antibodies (Dianova), respectively. As a loading control, the expression of β-actin was detected with anti-β-actin antibodies (Sigma Aldrich).
For the analysis of the activation of SARS-S and MERS-S for virus-cell fusion, we employed a previously described retroviral pseudotyping system [
To determine whether the TMPRSS2 isoforms can activate FLUAV, 293T cells were seeded in 6-well plates at a density of 2.8 x105 cells/well and cultured for 24 h. Then, the cells were transiently transfected with 6 μg of plasmids encoding the proteases or empty plasmid, which served as a control, using the calcium phosphate transfection method. After overnight incubation, the medium was replaced by fresh DMEM. At 24 h post transfection, the cells were gently washed with PBS and then incubated with DPBS with Mg2+ and Ca2+ (PAN Biotech) supplemented with 0.2% bovine serum albumine (BSA) (MACS Miltenyi Biotec) containing FLUAV A/PR/8/34 or A/Panama/2007/99 at a multiplicity of infection (MOI) of 0.01 or 0.1, respectively. After 1 h of incubation at 37°C in a humidified atmosphere, the cells were gently washed with PBS, and fresh infection medium (DMEM supplemented with 0.2% BSA, penicillin and streptomycin) was added. To analyze virus release, the culture supernatants were collected at 48 h post infection. The amount of infectious units within the culture supernatants was determined by focus formation assay, as described above.
To analyze the cellular localization of TMPRSS2 isoforms, COS-7 cells were transiently transfected with 6 μg of plasmids encoding TMPRSS2 isoforms or empty plasmid as control. After overnight incubation, the transfection medium was replaced with fresh DMEM. At 24 h post transfection, the cells were washed with PBS, and incubated for 1 h with DPBS with Ca2+ and Mg2+, supplemented with 0.2% BSA, containing FLUAV at an MOI of 0.5. Thereafter, the cells were washed with PBS and fresh infection medium was added. At 24 h post infection, the cells were fixed with ice cold methanol, blocked with 3% BSA for 1 h, and then stained with mouse anti-TMPRSS2 and rabbit-anti-PR8 HA antibodies (Santa Cruz and Sino Biological, respectively). After 1 h of incubation with primary antibodies, the cells were washed three times with PBS, and incubated for 1 h with anti-mouse and anti-rabbit secondary antibodies, coupled to Rhodamine Red-X and FITC (Dianova), respectively. After three final washing steps, the cells were stained with Vectashield mounting medium (Vector Laboratories) and analyzed with a Zeiss LSM 5 laser scanning microscope. Image capture was performed with Pascal Software (Zeiss) and Fiji software was used for image analysis and calculation of Pearson Correlation Coefficient.
Alternative splicing of the tmprss2 transcript is believed to generate mRNAs encoding for at least two different isoforms of the protein, isoforms 1 and 2 [
(A) Sequence alignment of the N-termini of TMPRSS2 isoforms 1 and 2. Identical amino acids are marked with stars. Amino acids absent in isoform 2 are marked with ‘–‘. (B) Plasmids encoding TMPRSS2 isoform 1 and isoform 2, both equipped with an N-terminal myc tag, were transiently transfected into 293T cells. Empty plasmid (pCAGGS) served as a negative control. Protease expression in cell lysates was detected via Western blotting with anti-myc antibody. The β-actin expression served as a loading control. Black-filled arrowheads indicate the zymogen forms, while grey-filled arrowheads highlight cleavage products resulting from protease activation. (C) 293T cells were transfected as described for panel B but protease expression was determined using flow cytometry with an anti-TMPRSS2 antibody. The geometric mean channel fluorescence (GMCF) measured in a representative experiment performed with triplicate samples is shown. Error bars indicate standard deviations. Similar results were obtained in two independent experiments.
We next determined whether transcript variant 1 (which encodes isoform 1) is expressed in tissues and cell lines. For this, we employed RT-PCR with primer sets specific for tmprss2 transcript variant 1 and 2, respectively. Primers specific for variant 1 amplified their target sequence from a plasmid encoding isoform 1 but not isoform 2, confirming that the PCR was specific. Similarly, the PCR designed to amplify transcript variant 2 was negative when a plasmid encoding isoform 1 was used as template but generated the expected amplificate from cDNA prepared from several cell lines and tissues, indicating that also this PCR was specific. In this context, it should be noted that the forward primer used for detection of transcript variant 2 binds to a 5’-untranslated region which is absent from the plasmid encoding isoform 2. Therefore, the respective PCR was negative.
The investigation of tissue samples revealed robust expression of transcript variant 1 in lung, kidney, liver and pancreas, which were reported to express tmprss2 mRNA (isoforms were not discriminated in previous studies) [
RT-PCR analysis of the expression of tmprss2 transcript variants 1 and 2 in organ samples from adult men (left panel) or from cell lines of human origin (right panel). Expression of GAPDH was assessed in parallel. Plasmids encoding isoform 1 and 2 (left panel) or a mix of both plasmids (right panel) served as positive control. Similar results were obtained in two separate experiments.
Activation of FLUAV by TMPRSS2 in infected cells requires that HA comes into contact with the protease. In order to address whether HA and TMPRSS2 isoform 1 colocalize, immunofluorescence staining and Fiji analysis of COS-7 cells transiently expressing TMPRSS2 isoforms and infected with FLUAV was employed. We observed colocalization of HA with both TMPRSS2 isoforms in infected cells (
(A) COS-7 cells were transfected with plasmids encoding TMPRSS2 isoform 1 or isoform 2 or with empty plasmid which served as negative control. Subsequently, the cells were infected with FLUAV A/PR/8/34 (H1N1) at an MOI 0.5. At 24 h post infection, the cells were stained for FLUAV-HA (green) and TMPRSS2 isoforms (red) and images were taken at 63 x magnification. White squares show examples of colocalization of HA and TMPRSS2 (yellow signals) and were digitally magnified 2.5x from the original images. Similar results were obtained in three separate experiments. (B) Images obtained in (A) were analyzed with Fiji software, which allows calculation of the Pearson Correlation Coefficient (PCC), a measure for colocalization. The average PCC measured in three separate experiments is shown. For each experiment, 6–8 cells were analyzed. Error bars indicate standard error of the mean (SEM).
In order to determine whether TMPRSS2 isoform 1 can activate FLUAV, we first assessed cleavage of FLUAV HA. For this purpose, we coexpressed the HA proteins of FLUAV A/South Carolina/1/1918 (H1N1) or FLUAV A/Hong Kong/1/1968 (H3N2) and proteases. Cleavage of the HA precursor, HA0, was observed upon treatment of cells with trypsin and upon coexpression of DESC1 and TMPRSS2 isoform 2, in keeping with previous reports [
(A) Expression plasmids encoding FLUAV HA subtypes H1 (left) and H3 (right) and the indicated proteases or empty plasmid (pCAGGS) were transiently cotransfected into 293T cells. At 48 h post transfection the cells were treated with PBS or trypsin, and HA cleavage was determined by Western blotting. Similar results were obtained in three independent experiments. The HA0 precursor (upper arrow) and the HA1 (middle arrow) and HA2 (lower arrow) subunits are indicated. (B) The indicated proteases were transiently expressed in 293T cells and the cells infected with FLUAV A/PR/8/34 (H1N1) at an MOI 0.01 (left) or FLUAV A/Panama/2007/99 (H3N2) at an MOI of 0.1 (right) and treated with either trypsin or PBS. At 48 h post infection, the virus release was measured by determination of infectious particles (ffu/ml) in the culture supernatant. The results of representative experiments performed with triplicate samples are shown. Error bars indicate standard deviations. Similar results were obtained in three independent experiments. ffu, focus forming units.
We next asked whether cleavage results in HA activation. To this end, we determined the spread of FLUAV A/PR/8/34 (H1N1) and FLUAV A/Panama/2007/99 (H3N2) in 293T cells transfected to express the indicated proteases. Treatment of cells with PBS or expression of TMPRSS3 did not promote viral spread (
Apart from FLUAV, several respiratory viruses hijack TMPRSS2 to facilitate activation of their envelope glycoprotein, including the emerging SARS-CoV [
(A) 293T cells were cotransfected with plasmid encoding SARS-S with a C-terminal V5 tag and plasmids encoding the indicated proteases. At 48 h post transfection, the cells were treated with PBS or trypsin, and SARS-S cleavage was analyzed by Western blotting with a V5-specific antibody. The expression of β-actin served as a loading control. The results are representative of three independent experiments with different plasmid preparations. Black-filled arrowhead, uncleaved SARS-S; gray-filled arrowhead, S2 subunit generated by trypsin digest; white-filled arrowheads, C-terminal cleavage fragments generated by TMPRSS2. (B) To analyze SARS-S-driven virus-cell fusion, 293T cells were transiently transfected with plasmids encoding the indicated proteases and ACE2. At 24 h post transfection, the cells were pretreated with medium supplemented with DMSO or 10 μM cathepsin B/L inhibitor MDL 28170, and transduced with pseudotypes bearing SARS-S. The luciferase activities in cell lysates were analyzed at 72 h post transduction. The results of a representative experiments performed with triplicate samples are shown. Error bars indicate standard deviations. Similar results were obtained in two separate experiments. (C) MERS-S-driven virus-cell fusion was analyzed as described for panel B but 293T cells transfected to express DPP4 were used as targets. The results of a representative experiment performed with triplicate samples are shown. Error bars indicate standard deviations. Similar results were obtained in three independent experiments. c.p.s., counts per second.
Several host cell proteases can activate FLUAV HA in cell culture and it has thus been assumed that respiratory viruses can employ redundant proteolytic systems to ensure their activation in the host [
The
An RT-PCR designed for the amplification of a sequence specific to transcript variant 1 showed that this transcript is expressed in cell lines and tissues susceptible to FLUAV infection and previously found to be positive for tmprss2 mRNA and or protein, including lung and liver [
Our mRNA expression data suggest that at least a fraction of the TMPRSS2 molecules present in FLUAV target cells correspond to isoform 1, raising the question whether they contribute to viral activation. Our study provides an affirmative answer: Both isoform 1 and 2 colocalized with HA in transfected cells, cleaved HA upon coexpression and activated authentic FLUAV of different subtypes. Moreover, isoform 1, like isoform 2, activated SARS-S and MERS-S for entry into target cells. Thus, the presence of an extended N-terminus does not seem to alter the ability of TMPRSS2 to activate respiratory viruses, although one should take into account that transfected cells might not mirror all aspects of cells endogenously expressing isoform 1.
Collectively, our results indicate that isoform 1 of TMPRSS2 can be produced in viral target cells and likely contributes to viral activation. Therefore, approaches to suppress HA activation for influenza therapy must target both isoforms and our initial results obtained with camostat mesylate indicate that this is feasible. However, it remains to be determined whether subtle differences in cellular localization and autocatalytic activation of TMPRSS2 isoforms observed here impact sensitivity to protease inhibitors in cells endogenously expressing these isoforms.
We thank M. Winkler for helpful discussion.