Alveolar epithelial cell injury with Epstein-Barr virus upregulates TGFβ1 expression

Andrea P. Malizia, Dominic T. Keating, Sinead M. Smith, Dermot Walls, Peter P. Doran, Jim J. Egan


Idiopathic pulmonary fibrosis (IPF) is a refractory and lethal interstitial lung disease characterized by alveolar epithelial cells apoptosis, fibroblast proliferation, and ECM protein deposition. Epstein-Barr virus (EBV) has previously been localized to alveolar epithelial cells of IPF patients and is associated with a poor prognosis. In this study, we utilized a microarray-based differential gene expression analysis strategy to identify molecular drivers of EBV-associated lung fibrosis. Two cell lines, primary human alveolar epithelial cells type 2 and A549 cells, were infected with EBV. EBV lytic phase induction increased active and total transforming growth factor-β1 (TGFβ1) transcript expression in association with reduced cell proliferation and increased caspase 3/7 activity. Exposing EBV-infected cells to ganciclovir resulted in TGFβ1 deregulation and reduced expression of EBV early response genes, BRLF1 and BZLF1. We targeted the BRLF1 and BZLF1 gene products, Rta and Zta, by silencing RNA, and this resulted in the normalization of TGFβ1 transcript and cell proliferation levels. Our study using a viral cell line model complements existing human and animal model data and further provides evidence to suggest that viral epithelial cell injury may play a role in IPF.

  • idiopathic pulmonary fibrosis
  • transforming growth factor-β1
  • epithelial injury
  • ganciclovir

idiopathic pulmonary fibrosis (IPF) is a lethal interstitial lung disease characterized histologically by fibroblastic foci, which underlie areas of unresolved epithelial injury and are sites where activated fibroblasts and myofibroblasts migrate, proliferate, and synthesize ECM proteins. The emerging hypothesis underlying the pathogenesis of IPF is that repeated microscopic injury of alveolar epithelial cells results in a fibroproliferative response, mediated by fibroblastic foci (36). The source of epithelial cell injury has received little attention, however, it is probable that no single mechanism initiates the disease, but rather a combination of injuries leads to the disease process.

Endogenous herpes viruses may be one important source of injury. Whereas Epstein-Barr virus (EBV) usually infects the upper respiratory tract, it has also been shown to infect and replicate in the lower respiratory tract (25). Both EBV protein and DNA expression have been identified in the lung tissue of IPF patients (11, 37). In surgically acquired lung tissue EBV glycoprotein-340 (gp340)/gp220 and VCA, viral proteins expressed during the lytic phase of EBV infection, have been localized to alveolar epithelial cells (11). Tang et al. (38) identified EBV infection detected by real-time PCR in 62.5% of familial IPF cases and 64% of sporadic IPF cases. The putative role of EBV in the development of IPF has been expanded by the observation that the expression of EBV latent membrane protein 1 (LMP1) in alveolar epithelial cells is associated with a poor prognosis (39). Further studies have suggested a pathogenic role for herpes virus infection by demonstrating clinical stability in two IPF patients following antiviral therapy (38).

The EBV BZLF1 and BRLF1 genes encode transactivators and activate viral/cellular promoters, which lead to an ordered cascade of viral gene expression (18). The BZLF1 gene product, Zta, is a transcriptional activator that binds to activator protein-1 (AP-1)-like sequences, whereas the BRLF1 gene product, Rta, binds directly to a GC-rich motif (7, 15, 16). Zta alone is sufficient to activate the EBV lytic cascade and is central to the regulation of cellular proliferation (3, 4, 26). Furthermore, these EBV immediate-early proteins, Zta and Rta, disrupt latency in EBV-infected cells, allowing the virus to form new particles, which are then transferred to new host cells (32, 41). The induction of lytic EBV infection allows the infected host cell to phosphorylate the antiviral prodrug ganciclovir into its cytotoxic form (28). This switch to a replicative pattern of viral gene expression can be mimicked by treating latently infected cells with phorbol ester and sodium butyrate (TPA/BA) or with transforming growth factor-β1 (TGFβ1) (8, 13).

The aim of this study was 1) to investigate the impact of infecting epithelial cells with EBV, 2) to identify a possible role for EBV infection in the expression of TGFβ1, and 3) to determine whether inhibiting EBV lytic phase infection induces alterations in TGFβ1 expression and cell activity.


Cell cultures and EBV infection in vitro.

Human primary alveolar epithelial cells were purchased from ScienCell (San Diego, CA) as a cell culture containing both type 2 (AEC2) and type 1 (AEC1) cells (14). Human primary alveolar epithelial cells are isolated from human lung tissue and cryo-served at primary culture and delivered frozen. AEC2 cells were grown and maintained in AEpiC medium (ScienCell) supplemented with 2% FCS and 10 ng/ml keratinocyte growth factor (Sigma) (40). AEC2 have a distinct morphology, appearing as large, somewhat square cells with characteristic lamellar bodies that contain surfactant lipids (27, 33).

Human alveolar epithelial cells (A549) were obtained from European Collection of Cell Cultures (Salisbury, United Kingdom) and grown in vitro in RPMI 1640 supplemented with 5% FCS and 146 mg/l l-glutamine (Sigma-Aldrich). B lymphocytes infected with EBV were obtained from Prof. Kenzo Takada, Hokkaido University School of Medicine, Sapporo, Japan. To effect latent virus infection, both epithelial cells lines were independently cocultured with the infected B cell line as previously described (19, 22). Briefly, infected B lymphocytes were maintained in RPMI 1640 supplemented with 10% FBS, 5% l-glutamine, 1% penicillin, and 1% streptomycin. A viral lytic cycle was induced in the cells by adding goat anti-human serum immunoglobulin G (Sigma). After 4 h, the B lymphocytes were added to both alveolar epithelial cell lines at a concentration of 5 × 105 cells/ml. Following 4 days of coincubation, the media containing B cells was removed, pelleted down, and resuspended in M5 media for 48 h (calcium-free DMEM supplemented with 5% horse serum, 2 mM glutamine, cortisol, 2 ng/ml EGF, 10 mg/ml insulin, 100 ng/ml cholera toxin, 1% penicillin, and 1% streptomycin). The alveolar epithelial cells were washed with PBS, and the media replaced containing 10% FCS. G418, 500 μg/ml and 200 μg/ml (Sigma) have been added for 10 days to select out A549 and AEC2 cells, respectively, that had been successfully transfected with EBV. A549-infected cells were labeled as VAAK cells, whereas AEC2 were labeled as V-AEC2. To avoid any implications of temporary cytotoxic effect, we utilized the EBV-infected cell line VAAK after several passages in normal RPMI 1640 containing 5% FCS.

To confirm the absence or presence of B cells contamination in VAAK and V-AEC2 cultures, we compared the expression of the B cell-specific CD19 and latent EBV LMP1 gene expression in A549, AEC2, VAAK, and V-AEC2 cultures. CD19 is a specific B cell antigen that is not expressed in epithelial cells. By using melting curve analysis, we confirmed the absence of contamination of B cells expressing CD19 mRNA in VAAK and V-AECs.

Quantitative real-time PCR.

cDNA level was assayed by using a Rotor-Gene 3.0 real-time PCR instrument (Corbett Research) and QuantiTect SYBR Green PCR Kit (Qiagen). Quantitative real-time PCR was performed with primers for TGFβ1, CD19, BRLF1, BZLF1, LMP1, and GAPDH (Table 1). Products were measured by absolute quantification and reported as a function of cycle threshold (Ct), the cycle number at which PCR amplification becomes linear. mRNA expression was normalized to control and GAPDH expression thus obtaining mean fold change values or ΔΔCt. Following cycling, to ensure specificity, melt curve analysis was carried out to verify the amplification of PCR products starting at 65°C and ramping to 95°C at 0.1°C/s. One peak in the melt curve ± 2°C indicated nonsecondary products were formed. Fold differences were calculated over control for each exposed group using normalized Ct values. Results were examined by GraphPad Prism version 4.00 (GraphPad Software, San Diego, CA).

View this table:
Table 1.

List of primers for quantitative real-time-PCR

Immunofluorescence staining.

Control A549 and VAAK cells were plated into Lab Tek Chamber Slide (Sigma-Aldrich) at a density of 2 × 104 cells/well and grown to ∼80% confluence. Cells were washed, fixed in ice-cold methanol for 10 min, and then soaked in PBS containing 0.1% BSA and 20% goat serum (Sigma-Aldrich) for 30 min to reduced secondary nonspecific bindings. Cells were exposed to mouse anti-EBV Rta 1:30 for 1 h (Argene), whereas controls were exposed to PBS, 0.1% BSA, and 20% goat serum in absence of anti-EBV Rta antibody. After washing, bound primary antibody was detected using appropriate goat anti-mouse Texas red-conjugated secondary antibody (1:100; Santa Cruz Biotechnology). Images were collected using a ProgRes camera microscope system (Jenoptik Laser).

Lytic phase induction and inhibition.

VAAK and V-AEC2 cells were exposed in different experiments to 30 ng/ml 12-O-tetradecanoylphorbol 13-acetate (TPA), 3 mM sodium butyrate (BA), 15 μM curcumin, 10 μg/ml ganciclovir, and 10 μM SB-431542 at several time points. Curcumin has been utilized as an inducer of apoptosis by reducing AP-1 transcription factor activity in EBV-infected cells, as previously described (17). We used SB-431542, an inhibitor of ALK4/5/7 via TGFβRI, to inhibit autocrine TGFβ1-induced apoptosis, as previously described (20). All reagents were purchased from Sigma-Aldrich.

Cell proliferation assay.

The CellTiter 96 AQueous Non-Radioactive Cell Proliferation Assay (Promega) is a MTS reagent-based colorimetric method for determining the number of viable cells in proliferation. Cells were seeded in a 96-well plate at 5 × 103 cells per 100 μl of culture media per well for 24 h, MTS was added for 2 h, and then colorimetric signals from each condition were measured at 0, 6, 12, 24, 36, and 48 h. The quantity of formazan product as measured by the amount of 450-nm absorbance was directly proportional to the number of living cells in culture. Cell proliferation data were normalized to cell count by crystal violet staining. Crystal violet staining, which binds cell nuclei, gives a 595-nm optical density (OD) reading that is proportional to cell number (31).

Caspase 3/7 activity assay.

To measure apoptosis, we estimated caspase 3/7 activity by using the chemiluminescent Caspase-Glo 3/7 Assay (Promega). The assay provides a luminogenic caspase-3/7 substrate, which contains the tetrapeptide sequence DEVD, in a reagent optimized for caspase activity, luciferase activity, and cell lysis. Cells were grown in a 96-well plate, 100 μl of reagent were added to each well, and then the plates were incubated for 3 h at room temperature. Relative light units (RLU) were measured by using a luminometer and normalized to cell count by crystal violet staining.

Microarray analysis.

RNA isolation, cDNA synthesis, in vitro transcription, and microarray analysis were performed as previously reported (23). Image files were obtained through Affymetrix GeneChip software (MAS5); subsequently, robust multichip analysis (RMA) was performed. All samples were microarrayed in duplicate, an average RMA value was computed to ensure that the average was statistically representative, and P values were generated. Only those genes with a P value <0.01 were included in subsequent bioinformatic analysis. Expression data for each experiment were compared with control, and a signal log ratio (SLR) of 0.6 or greater (equivalent to a fold change in expression of 1.5 or greater) was taken to identify significant differential regulation (1). All the SLRs data resulting from comparative analyses were reported in a graph to determine the reliability of the assay and linearity by r2. For all the microarray assays, r2 values were >0.98. Using gene expression values normalized by RMA, average linkage hierarchical cluster analysis was performed, and the results were visualized by TreeView software (12). A list of dysregulated gene was curated via the publicly available DAVID 2.1 Bioinformatic Resources database (9, 18a).

TGFβ1 protein assay.

Supernatants were collected at several time points and stored at −80°C. Active and total TGFβ1 protein concentration was determined by TGFβ1-ELISA kit (PromoKine) following standard protocol. TGFβ1 protein concentration in the supernatant was normalized to cell count by crystal violet staining (31).

Generation and transfection of siRNA.

BRLF1 sequence (NCBI acc. no. NC_007605) was used to determine small interfering RNA (siRNA) target sequences using siRNA design tool web-based criteria (Qiagen). siRta544 20-nmol duplex sequence (AAG GCC TCC TAA GCT CCA AGT) and siRta276 20-nmol duplex sequence (AAT GCC GGC AAC GAC GCA TGT) (Qiagen) were named according to their location in BRLF1 genome. The siZta duplex sequence was used to inhibit BZLF1 expression as described previously (5). Control cells were transfected with nonsilencing siRNA duplexes (Qiagen). VAAK cells were plated in T-25 flasks for 48 h and then transfected with 6.25 μg of siRNA using RNAiFect (Qiagen).

After 24-h incubation with siRta544, siRta276, siZta, or siControl duplexes, VAAK cells were exposed to lytic phase induction by TPA/BA, and assays were performed at different time points. Total RNA was harvested at various times posttransfection by using TRIzol reagent (Sigma) as previously described (6).

Statistical analysis.

The statistical significance of our experimental findings was analyzed using Pearson analysis and F test two-sample for variances for cell proliferation data and Student's t-test assuming unequal variances was used for quantitative RT-PCR, TGFβ1 protein, and caspase 3/7 assays. Data with P < 0.05 are considered significant. The data shown are the representative average of at least two independent experiments performed in parallel except where stated otherwise in the text. All data are reported with their relative SD data (mean ± SD) except for caspase 3/7 luminescence assay data, which are reported with relative SE.


Impact of EBV infection on cell proliferation.

LMP1 was found to be expressed in both EBV-infected VAAK and V-AEC2 but not in control A549 cells or AEC2. EBV Rta protein expression was localized in A549 and VAAK cells, confirming nuclear protein localization by immunofluorescence (Fig. 1, A and B).

Fig. 1.

Validation of Epstein-Barr virus (EBV) infection in alveolar epithelial cells. A: latent membrane protein 1 (LMP1) and CD19 melting curve in control A549 (red), EBV-infected A549 (VAAK; black), EBV-infected alveolar epithelial type 2 (V-AEC2; blue), control epithelial AEC2 (gray), and EBV-infected Akata cells (B cells; green) by real-time quantitative PCR (qPCR). B: EBV early transcript Rta localization and expression in A549 and VAAK cells. Immersion oil magnification ×100. Tm, temperature of melting; dF/dT, fluorescence.

There were no significant differences in VAAK cell proliferation following EBV latent infection compared with control A549 cells (r = 0.989, F = 1.75, P = 0.275; Fig. 2A). Apoptosis was not significantly altered in any cell line, control A549 (3,496 ± 1,000 RLU/cell), control AEC2 (3,303 ± 950 RLU/cells), VAAK (5,000 ± 1,200 RLU/cells), and V-AEC2 cells (4,200 ± 1,260 RLU/cells) (Fig. 2B).

Fig. 2.

Cell proliferation following EBV infection in conjunction with exposure to EBV lytic phase inducer 12-O-tetradecanoylphorbol 13-acetate and 3 mM sodium butyrate (TPA/BA) and apoptosis inducer curcumin. A: cell proliferation data of A549 (−) and VAAK (X) cells. n = 4, r = 0.989, F = 1.75, P = 0.275. The error bars represent SD. B: apoptosis measured by caspase 3/7 activity in control A549, VAAK, control AEC2, and V-AEC2 at basal conditions. n = 4; Student's t-test, P = 0.5. The error bars represent SE. C: VAAK cell proliferation data following TPA/BA lytic phase induction and curcumin apoptosis induction in EBV-infected cells at different time points. VAAK cells and TPA/BA: n = 4, r = 0.896, F = 20.41, *P = 0.0466. VAAK cells and curcumin: n = 4, r = −0.106, F = 33.71, *P = 0.0288. D: caspase 3/7 activity measured in VAAK and V-AEC2 after 24-h TPA/BA lytic phase induction and curcumin exposure. Student's t-test, *P < 0.05. O.D., outer diameter; RLU, relative light units; T, time.

Following EBV lytic phase induction by TPA/BA and curcumin apoptosis induction, VAAK cell proliferation decreased at different time points compared with latent controls (n = 4; TPA/BA: r = 0.896, F = 20.41; curcumin: r = −0.106, F = 33.71; P < 0.05; Fig. 2C). EBV lytic phase induction by TPA/BA increased apoptosis measured by caspase 3/7 activity in VAAK cells and V-AEC2 compared with control (VAAK, n = 4; TPA/BA: 9,953 ± 1,210 RLU/cells; P < 0.05; curcumin: 7,982 ± 1,315 RLU/cells; V-AEC2, n = 4; TPA/BA: 8,144 ± 1,260 RLU/cells; P < 0.05; curcumin: 6,000 ± 980 RLU/cells; Fig. 2D).

Altered gene expression following latent EBV infection.

EBV infection of A549 cells was associated with significant changes in gene expression determined by oligonucleotide array analysis (GEO acc. no. GSE5450). Of the 22,216 gene sequences represented on the Affymetrix HGU133A oligonucleotide array, 1,551 genes were significantly altered following EBV infection (Fig. 3A). Significantly altered genes were used as input in functional classification searches by GO charts analysis. Of note with respect to the pathogenesis of IPF was the finding of significantly altered expression of genes in the cell proliferation group. This group was composed of 95 altered transcripts, including 30 cell proliferation related genes that were associated with negative regulation (31%) (Table 2).

Fig. 3.

Hierarchical clustering of dysregulated probe sets. A: overall unsupervised hierarchical cluster of dysregulated gene in each duplicated microarray analysis along with the average signal log ratio (SLR). SLR expressions are reported in relation to their yellow-blue color-based intensity. Groups of altered genes are found to cluster together, and in particular the gene expression profiles of each cell line have a high similarity, as has been reported at the right of the dendogram by Pearson analysis. B: transforming growth factor-β1 (TGFβ1) probe set SLRs expression in A549 and VAAK microarray analysis. Clustering analysis reports a significant correlation grouping A549 and VAAK microarray analysis along with their average signals.

View this table:
Table 2.

List of probe sets related to altered genes in A549 after EBV infection associated with negative regulation of cell proliferation functional group

The altered expression of cell proliferation correlated transcripts and the cytokines-mediated apoptosis pathways (35) included TGFβ1 upregulation in latent EBV-infected A549 (SLR 0.65) (Fig. 3B). Although gene expression is associated with the yellow-blue color-based intensity, adjusting array data set by Gene Cluster 3.0 software may alter the presentation of data intensity of gene expression for both control A549 and VAAK cells (Fig. 3B; Ref. 12). Microarray data of control A549 and VAAK cells treated with 10 ng/ml TGFβ1 for 4 h showed increased TGFβ1 transcript expression in VAAK cells compared with control A549 cells (GEO acc. no. GSE5457; A. P. Malizia, D. Walls, J. J. Egan, and P. P. Doran, unpublished observations).

TGFβ1 expression in response to latent and lytic infection of EBV.

TGFβ1 mRNA expression increased 2.4- and 3.0-fold in latently infected VAAK cells and V-AEC2 compared with control epithelial cells at basal conditions (n = 4, ΔΔCt 2.4 ± 0.2, ΔΔCt 3.0 ± 0.5, respectively; P < 0.05, P < 0.01; Fig. 4A). Active and total TGFβ1 protein levels were found to be higher in VAAK cells (n = 4, 6.01 ± 0.3 ng/ml, 7.3 ± 0.4 ng/ml, respectively) than in control A549 at basal conditions (n = 4, 2.4 ± 0.1 ng/ml, 3.38 ± 0.2 ng/ml, respectively; P < 0.05; Fig. 4B).

Fig. 4.

TGFβ1 and EBV early response genes BRLF1 and BZLF1 expression in EBV-infected epithelial cells. A: TGFβ1 RNA expression in A549, VAAK, AEC2, and EBV-infected AEC2. Student's t-test, *P < 0.05, **P < 0.01. The error bars represent SD. B: TGFβ1 protein concentration (conc.) in A549 and VAAK supernatant in conjunction with EBV lytic phase and apoptosis. n = 4; Student's t-test, *P < 0.05, **P < 0.01. C: TGFβ1 protein concentration in AEC2 and V-AEC2 supernatant along with EBV lytic phase activation and apoptosis induction. n = 4; Student's t-test, *P < 0.05, **P < 0.0. D: by real-time qPCR, BRLF1 expression was determined in VAAK cells in response to lytic phase induction by TPA/BA and antiviral ganciclovir (GCV) exposure. n = 4; Student's t-test, *P < 0.05. E: BZLF1 expression in VAAK cells determined by real-time qPCR following lytic phase induction and ganciclovir exposure. n = 4; Student's t-test, *P < 0.05.

Following EBV lytic phase induction, total and active TGFβ1 protein concentration were found to be higher in VAAK supernatant (n = 4, active and total TGFβ1, 8.12 ± 0.4 ng/ml, 14.4 ± 1.6 ng/ml) and after curcumin apoptosis induction for 24 h (n = 4, active and total TGFβ1, 9.51 ± 0.5 ng/ml, 12.8 ± 1.6 ng/ml) compared with control A549 and VAAK at basal conditions (P < 0.01; Fig. 4B).

There was also a significant increase of active and total TGFβ1 protein in V-AEC2 (n = 4, active and total TGFβ1, 3.625 ± 0.18 ng/ml, 6.975 ± 0.4 ng/ml) compared with control AECs at basal conditions (n = 4, active and total TGFβ1, 1.60 ± 0.11 ng/ml, 4.11 ± 0.25 ng/ml; P < 0.05) (Fig. 4C). TPA/BA and curcumin exposure induced a significant increase in active and total TGFβ1 protein expression after 24 h compared with control AEC (n = 4; P < 0.05, P < 0.01; Fig. 4C).

Ganciclovir reduces EBV lytic phase infection in VAAK.

Following lytic phase induction, BRLF1 mRNA expression increased by 2.4-fold (ΔΔCt 2.4 ± 0.15; P < 0.05), whereas BZLF1 mRNA expression increased by 1.8-fold (ΔΔCt 1.8 ± 0.1; P < 0.05) when compared with controls (Fig. 4, D and E).

VAAK cells treated with ganciclovir had both reduced BRLF1 mRNA expression (67%, ΔΔCt 0.8 ± 0.05; P < 0.05) and reduced BZLF1 mRNA expression (12%, ΔΔCt 1.6 ± 0.05; P = 0.5) (Fig. 4, D and E).

Ganciclovir reduces TGFβ1 expression and apoptosis.

Induction of EBV lytic phase infection by TPA/BA exposure led to a significant increase in TGFβ1 mRNA in both VAAK and V-AEC2 cells (VAAK ΔΔCt 2.7 ± 0.15, V-AEC2 ΔΔCt 3.5 ± 0.3; P < 0.05, P < 0.01, respectively), which was suppressed by ganciclovir (Fig. 5A).

Fig. 5.

Ganciclovir reduces TGFβ1 expression and increases cell proliferation. A: TGFβ1 mRNA expression levels determined by real-time qPCR in VAAK and EBV-infected AEC2 and following TPA/BA lytic phase induction and then treatment by ganciclovir (GCV). n = 4; Student's t-test, *P < 0.05, **P < 0.01. B: active and total TGFβ1 concentrations in VAAK supernatant, normalized to nuclear crystal violet staining (595 nm OD), following treatment with ganciclovir and an inhibitor of TGFβ superfamily type I activin receptor-like kinase receptors ALK4/5/7, SB-431542. n = 4; Student's t-test, *P < 0.05, **P < 0.01. C: TGFβ1 protein concentration in TPA/BA pretreated EBV-infected AEC2 supernatant in conjunction with ganciclovir (GCV) and SB-431542. n = 4; Student's t-test, *P < 0.05. D: cell proliferation assay data normalized to cell count. VAAK (▴), VAAK and ganciclovir (−), VAAK and lytic phase induction by TPA/BA (□), VAAK, TPA/BA, and ganciclovir (▪). n = 4, r = 0.863, F = 31.2, ***P < 0.001. n = 4, r = 0.981, F = 6.07, *P < 0.05. E: caspase 3/7 activity in VAAK and EBV-infected AEC2 in response to ganciclovir and to SB-431542 following lytic phase induction by TPA/BA. n = 4; Student's t-test, *P < 0.05.

Following induction of EBV lytic phase infection, ganciclovir treatment resulted in a significant decrease in TGFβ1 mRNA expression (VAAK ΔΔCt 1.1 ± 0.06, V-AEC2 ΔΔCt 1.5 ± 0.2; P < 0.05; Fig. 5A) and decreased active and total TGFβ1 protein levels in both cell supernatants (n = 4; P < 0.05; Fig. 5, B and C). Furthermore, there was a reduction in active and total TGFβ1 protein concentration in VAAK cells and V-AEC2 exposed to TPA/BA and SB-431542 compared with TPA/BA exposure only (P < 0.05; Fig. 5, B and C).

VAAK cells and V-AEC2 were initially exposed to ganciclovir in isolation to determine any nonspecific effects in TGFβ1 expression. Neither TGFβ1 mRNA nor protein expression were altered after 24-h ganciclovir exposure. Ganciclovir itself had no cytotoxic or reduced cell activity effects in VAAK (r = 0.989, F = 1.03, P = 0.48), whereas lytic phase induction with TPA/BA exposure reduced significantly cell proliferation at several time points (r = 0.863, F = 31.2, P < 0.001). We found that ganciclovir treatment enhances VAAK cell proliferation (n = 4, r = 0.981, F = 6.07, P < 0.05; Fig. 5D). Caspase 3/7 activity after EBV lytic phase induction was reduced in response to ganciclovir treatment and SB-431542 exposure (Fig. 5E).

Silencing EBV immediate early genes Rta and Zta enhances cell survival.

Rta and Zta expression were inhibited in VAAK cells and V-AEC2 after EBV lytic phase induction by TPA/BA using silencing RNA technique, as previously described (5, 21).

siRta276 and siRta544 transfection resulted in significant decreases in BRLF1 expression by 78% and 80%, respectively (ΔΔCt 0.53 ± 0.02, P < 0.01; ΔΔCt 0.48 ± 0.025, P < 0.01; Fig. 6A).

Fig. 6.

Inhibiting EBV early genes BRLF1 and BZLF1 facilitates cell survival. A: BRLF1 expression in response to siRta544 and siRta276 duplexes transfection following lytic phase induction with TPA/BA. n = 4; Student's t-test, **P < 0.001. B: active and total TGFβ1 protein concentration in VAAK and EBV-infected AEC2 cultures 24 h following siControl, siRta, and siZta transfection following TPA/BA lytic phase induction. 595 nm OD. n = 4; Student's t-test, *P < 0.05. C: cell proliferation assay data normalized to cells count. siControl transfected cells and TPA/BA (□). siRta and TPA/BA (▪), n = 4, r = 0.21, F = 7.26, P = 0.12. siZta and TPA/BA (▴), n = 4, r = −0.67, F = 235, **P = 0.004. D: caspase 3/7 activity in VAAK and EBV-infected AEC2 supernatant in response to siRta and siZta duplexes transfection following lytic phase induction with TPA/BA, normalized to nuclear crystal violet staining (595 nm OD). n = 4; Student's t-test, P = 0.5.

To determine a potential role of EBV Rta and Zta expression in the regulation of TGFβ1 expression, we examined its transcription in VAAK cells and V-AEC2 after siRta544, siZta, and nonsilencing siControl transfection following EBV lytic phase induction by TPA/BA. Despite a decrease in both total and active TGFβ1 protein in siRNA transfected cell lines, only active TGFβ1 protein decreased significantly after siZta and siRta transfection compared with nonsilencing siControl transfected cells (P < 0.05; Fig. 6B).

Cell proliferation decreased significantly in nonsilencing siControl transfected cells exposed to EBV lytic phase induction by TPA/BA at 4 and 24 h (Fig. 6C). siRta and siZta also impaired cell proliferation following TPA/BA exposure (r = 0.21, F = 7.26, P = 0.12; r = −0.67, F = 235, P < 0.01, respectively), suggesting a link between reduced EBV lytic phase reactivation and reduced caspase 3/7 activity (Fig. 6, C and D).

EBV lytic phase-associated injury via TGFβ1 expression.

We examined TGFβ1 promoter sequence by computational biology to identify potential sites of regulation of EBV. None of the potential binding sites that match for Zta transcription factor were found in the promoter sequence gene by using Transcription Element Search System (TESS) search database (; Ref. 34). By using ElDorado and MatInspector web-based software (; Ref. 2), we found three possible transcription factor binding sites for Rta-EBV transcription factor R (V$REBV) in TGFβ1 gene promoter sequence, with a matrix similarity score higher than 0.80. These data suppose a possible role, as enhancer factor, of Rta in the interaction and transcription of TGFβ1.

To validate these data, we used EMSA technique investigating the transcription factor binding activity of nuclear proteins extracted from 1-, 4-, and 24-h TPA/BA-exposed cells.

EMSA gel electrophoresis of reaction mixture of nuclear proteins and oligonucleotides containing Rta transcription factor binding site (+) and non-sense binding site (−) showed no significant DNA/proteins complex signals in PAGE (data not shown).


The pathogenesis of IPF is thought to result from repetitive epithelial cell injury. The development of a fibrotic lesion may also result from an imbalance in apoptosis and impaired epithelial cell activity. Viral infection may be a source of epithelial cell injury, as both EBV antigens and EBV DNA have been localized to lung tissue of IPF patients. Our data indicate increased TGFβ1 expression in epithelial cells following EBV infection, which is reduced by ganciclovir therapy. Furthermore, there is reduced cell proliferation and increased apoptosis following EBV lytic phase infection. This imbalance may contribute to the tissue remodeling seen in IPF.

Recent data in an animal model emphasize that herpes viruses can contribute to the development of pulmonary fibrosis (24). Chronic murine γ-herpesvirus 68 (MHV68) infection in IFN-γ-deficient mice, localized to type II alveolar cells in a fashion similar to human tissue, has resulted in the deposition of collagen in the lung (29). The application of antiviral therapies in this model subsequently has been shown to arrest pulmonary fibrosis (30). This effect appears to have been mediated by reducing the burden of virus-infected B cells. Extending on those observations in animal models, we employed two novel virus-infected alveolar epithelial cell lines to evaluate the impact of EBV infection. EBV infection of alveolar epithelial cells induced significant alterations in gene expression. Of note with respect to the pathogenesis of IPF was the findings of significant changes in the enhanced expression of TGFβ1.

To further determine the molecular mechanisms of EBV-driven apoptosis, we employed a gene knockdown strategy. Using siRNA probes, we demonstrated an indirect role for the EBV early gene, Rta and Zta, in TGFβ1 deregulation. TGFβ1 mRNA and active protein were reduced during silencing RNA treatment. Furthermore, inhibition of the EBV replicative phase reduced TGFβ1 expression both directly by silencing mRNA strategy and with antiviral therapy. Ganciclovir, the antiviral drug that is effective against EBV, inhibited Zta and Rta expression in VAAK in response to lytic phase induction by TPA/BA exposure, suggesting epithelial cell survival after antiviral treatment.

Cell line models like those described may not accurately reflect in vivo mechanisms; therefore, we utilized two different cell lines to explore the impact of viral infection on these cells. Both epithelial cell lines provided comparable data. Although these cell lines are widely utilized, the model does not guarantee that this is representative of human lung tissue.

In conclusion, we have demonstrated that TGFβ1 is induced following latent EBV infection of epithelial cells, and EBV lytic phase induction may increase apoptosis of alveolar epithelial cells. Furthermore, EBV early gene Rta and Zta may indirectly regulate TGFβ1 expression, promoting an indirect injury associated with infection. Therefore, these data demonstrate the functional activity of EBV in modulating epithelial cell injury and repair. Our study utilizing a viral cell line model complements existing human and animal model data and provides evidence to suggest that viral epithelial cell injury may play a role in IPF.


We acknowledge the support of the European Union, the Irish Lung Foundation, and the Irish governments' Programme for Research in Third Level Institutions.


  • The costs of publication of this article were defrayed in part by the payment of page charges. The article must therefore be hereby marked “advertisement” in accordance with 18 U.S.C. Section 1734 solely to indicate this fact.


  1. 1.
  2. 2.
  3. 3.
  4. 4.
  5. 5.
  6. 6.
  7. 7.
  8. 8.
  9. 9.
  10. 11.
  11. 12.
  12. 13.
  13. 14.
  14. 15.
  15. 16.
  16. 17.
  17. 18.
  18. 18a.
  19. 19.
  20. 20.
  21. 21.
  22. 22.
  23. 23.
  24. 24.
  25. 25.
  26. 26.
  27. 27.
  28. 28.
  29. 29.
  30. 30.
  31. 31.
  32. 32.
  33. 33.
  34. 34.
  35. 35.
  36. 36.
  37. 37.
  38. 38.
  39. 39.
  40. 40.
  41. 41.
View Abstract