Primary normal HLF were purchased from Lonza (Walkersville, MD, USA) and maintained in fibroblast basal medium supplemented with cytokines and growth factors and 2% serum (FGM-2 bulletkit, Lonza). Primary HBE cells were from Lonza and maintained in bronchial epithelial basal medium supplemented with cytokines and growth factors (BEGM, Lonza). Cells (0.5–1 × 10⁶) were grown as monolayers and cultured until about 90–95% confluent, then treated with 0.05% solution of Trypsin/EDTA, washed with HBSS (Corning, Manassas, VA, USA), and resuspended in medium for experiments. Cells were used between passages 2–5 for HLF and 2–3 for HBE. Cells from several donors were used for reproducibility.

Trypsin/EDTA and antibiotic/antimycotic solution (penicillin/streptomycin, thereafter P/S) were from Life Technologies (Carlsbad, CA, USA). Fetal bovine serum was from Omega Scientific (Tarzana, CA, USA). Propidium iodide (PI) was from Sigma (St. Louis, MO, USA). Recombinant human LIGHT/TNFSF14, lymphotoxin α1/β2, TGF-β1, BTLA-Fc chimera protein, IL-13, and human antibody to MMP-9 were obtained from R&D Systems (Minneapolis, MN, USA). Antihuman CD270 (HVEM clone eBioHVEM-122) and CD54 (ICAM-1, clone HA58) were from eBioscience (San Diego, CA, USA). Anti-human CD106 (VCAM-1, clone STA) and LTβR (clone 31G4D8) were from BioLegend (San Diego, CA, USA). Antibodies to IL-4Rα and IL-13Rα were from R&D. Flow cytometry staining was as described previously (15). Isotype matched mAb were used as negative controls. Cells were acquired with a FACSCalibur (Becton Dickinson, Mountain View, CA, USA) and analyzed using FlowJo software.

Human lung fibroblast or HBE were seeded a day prior to starting the treatment in 6 well plates (Becton Dickinson, Mountain View, CA, USA) at a density of 0.05–0.25 × 10⁶/ml in 2 ml. Cells were cultured alone, or with recombinant LIGHT/TNFSF14 (6–100 ng/ml), recombinant lymphotoxin α1/β2 (50 ng/ml), BTLA-Fc (500 ng/ml), recombinant IL-13 (25 ng/ml), or TGF-β1 (0.1–5 ng/ml) for 1–3 days. Cells were generally cultured for 48 hrs for mRNA analysis, a time point found to be optimal for LIGHT induction of target genes. In some cases, cells were stimulated with a membrane form of LIGHT present on the surface of a transfected cell line (EL4-LIGHT). Fibroblasts stimulated with membrane-bound LIGHT displayed similar responses compared to those stimulated with soluble LIGHT (not shown). HBE were stimulated in submerged cultures based on initial experiments showing similar results in air–liquid interface cultures.

Fibroblasts were cultured for 3 days and proliferation was studied by two methods: (1) thymidine uptake and (2) evaluation of DNA content. For thymidine uptake, 0.5 μCi [3H]-TdR (specific activity, 5.0 Ci/mmol; PerkinElmer, Waltham, MA, USA) was added 16 h before the end of the culture, and cells were harvested on MicroBeta FilterMate semi-automatic cell harvester (PerkinElmer). The incorporated [3H]-TdR was measured in a Beckman liquid scintillation counter, and results were expressed as counts per minute. To evaluate DNA content, cells were permeabilized with ice-cold 70% ethanol for 10 min. Cells were then washed with PBS and stained for 30 min at 37°C with 50 µg/ml of PI. Labeled cells were immediately acquired in a FACScalibur and DNA content (PI fluorescence) monitored on the FL2 channel.

RNA isolation and real time PCR analysis was performed as before (15) using specific primers (IDT, Coralville, IA, USA) already described (15). PCR products were generated in triplicate and normalized to housekeeping genes, β-actin or GAPDH. Relative quantification (RQ) was derived from the difference in cycle threshold (Ct) between the gene of interest and the housekeeping genes using the equation RQ = 2^−ΔΔct. Data in each experiment are displayed as the average RQ or fold increase above baseline expression.

ON-TARGETplus siRNA were purchased from Dharmacon (Pittsburg, PA, USA). 5–25 nM of control, HVEM, or LTβR siRNA were transfected into fibroblasts using HiPerFect transfection reagent (Qiagen, Venlo, Netherlands) according to the manufacturer’s instructions. Knockdown was monitored by mRNA and surface protein levels and was stable during the 48hr culture period with LIGHT.

Supernatants from stimulated HLF or HBE were centrifuged at 13,000 g for 15 min and stored at −80°C until used. The levels of IL-6, GM-CSF, CCL5, CXCL11 and CXCL12, TSLP, IL-33, ADAM8, MMP-9, and CCL26 were determined with ELISA-based kits (R&D Systems), according to the manufacturer’s instructions. MMP-9 immunoblotting was performed as before (15). Briefly, supernatants of treated cells were precipitated with TCA, resuspended in sample buffer and resolved in polyacrylamide gels under reducing conditions, then transferred to PVDF membranes and incubated with anti-MMP-9 followed by horseradish peroxidase-conjugated secondary Ab.

Statistical analyses were performed using Excel or GraphPad Prism 6 software. A two-tail unpaired Student’s t-test between two groups, or one-way ANOVA with Greenhouse-Geisser correction followed by Dunnett’s multiple comparisons test, was used to test the significance of the differences between group means. All data are representative of at least three independent experiments with different donor cell populations and expressed as mean ± SEM. Statistical significance was defined as *P < 0.05.

We examined the expression of LTβR and HVEM, the receptors for LIGHT, in five normal HLF cultures obtained from individual donors purchased from Lonza. All HLF cells expressed constitutively high levels of LTβR. HVEM was also constitutively expressed but at markedly lower levels, confirmed by flow cytometry as well as PCR (Figures 1A,B). The pattern of expression of both receptors was similar between donors, suggesting that HLF could be highly responsive to LIGHT.

Our previous data from in vivo studies of murine models of severe asthma (11) and SSc (12) showed that LIGHT promoted an increase in lung smooth muscle mass, in a manner partially dependent on TGF-β. This might have been due to hyperplasia of mature smooth muscle as well as accumulation of fibroblasts expressing alpha smooth muscle actin. We then first tested whether LIGHT could promote the division of HLF, relevant to the latter activity. As shown in (Figure 1C), treatment with soluble recombinant LIGHT significantly enhanced uptake of tritiated thymidine by HLF as well as induced greater cell cycle progression into S and G2/M phases as measured by PI staining (Figure 1D). This response was dose dependent, with maximal activity at 25–100 ng/ml of the soluble molecule (Figure 1E).

Previous data have shown that TGF-β can act on fibroblasts. However, in contrast to the positive effect of LIGHT on HLF division, TGF-β blocked the proliferation of these cells when cultured alone as well as when cocultured with LIGHT (Figures 1C–E). TGF-β on the other hand has been reported to promote alpha smooth muscle actin expression in HLF (9, 17). We confirmed alpha smooth muscle actin induction by TGF-β but found that LIGHT did not significantly promote this feature alone, or enhance the activity of TGF-β. In fact, LIGHT had a moderate negative effect on the ability of TGF-β to promote alpha smooth muscle actin, although this somewhat dependent on the time analyzed or dose of TGF-β used (Figure 1F; Figure S1 in Supplementary Material). These results suggest that LIGHT and TGF-β have distinct actions, but may work together albeit at different times, to enhance the accumulation of greater numbers of lung fibroblasts that express alpha smooth muscle actin.

In previous studies, we found that LIGHT stimulation of HBE cells resulted in endocytosis or cleavage of LTβR from the membrane, indicative of active signaling through this receptor (15). Similarly, exposure of HLF to LIGHT resulted in downregulation of surface LTβR but not HVEM (data not shown), implying that LTβR might be the primary receptor active on these cells. To investigate the contributions of HVEM and LTβR to cell division, HLF were then transfected with siRNA against these molecules. Flow cytometry and mRNA expression confirmed strong downregulation of both receptors with no effect on the non-targeted receptor (Figures 2A,B). This revealed that LIGHT-induced proliferation was dependent on LTβR but not HVEM (Figures 2C,D).

We also reported that LIGHT can augment the expression of adhesion molecules in bronchial epithelial cells (15). As this type of activity might also be important for the accumulation of fibroblasts in the lungs, either affecting their migration within the lung and/or maintenance around the sub-bronchiolar epithelial regions, we examined whether LIGHT could promote expression of ICAM-1 and VCAM-1 in HLF. All HLF constitutively expressed high levels of VCAM-1 and lower levels of ICAM-1. Importantly, soluble LIGHT upregulated both VCAM-1 and ICAM-1, in a time-dependent manner, at both the mRNA and protein levels (Figures 3A,B). LIGHT can be naturally expressed by T cells and other immune cells as a cell surface molecule or soluble after being cleaved from the membrane. We therefore compared soluble recombinant LIGHT to a membrane form of LIGHT that cannot be cleaved that was presented on the surface of a transfected cell line (EL4-LIGHT). Showing a similar activity, membrane LIGHT also strongly enhanced expression of both VCAM-1 and ICAM-1 (Figure 3A).

As an alternative way to investigate the contributions of HVEM and LTβR to the LIGHT-induced response, we stimulated HLF with either LTαβ or BTLA-Fc that signal through LTβR and HVEM, respectively. Confirming the prior data, after LTαβ stimulation the levels of both ICAM-1 and VCAM-1 were similar to those in LIGHT-treated HLF at both protein and mRNA levels. In contrast, no upregulation was observed with BTLA-Fc even though BTLA-Fc was able to bind HVEM (Figures 3C,D and data not shown). Therefore, LIGHT promotes several activities relevant to accumulation of fibroblasts in the lungs, mediated by signaling through LTβR.

Fibroblasts can also produce chemokines and other proinflammatory mediators that contribute to lung inflammation and tissue remodeling in asthma and other similar diseases. We therefore assessed the contribution of LIGHT to several of these activities in HLF. Treatment with soluble LIGHT specifically enhanced gene expression of CCL5, and to a lesser extent CCL20, major chemotactic factors for T cells, eosinophils, and basophils (Figure 4A). In contrast, other CC chemokine genes, like CCL2, CCL3, CCL17, CCL21, and CCL22, chemoattractants for monocytes/macrophages, and CCL11, CCL24, and CCL26, chemoattractants for eosinophils, were not upregulated after LIGHT stimulation (data not shown). Also, we found that LIGHT enhanced transcription of the genes for CXCL5 and CXCL11, potent chemoattractants for lymphocytes and neutrophils, and CXCL12, that aids chemotaxis of neutrophils and lymphocytes and synergizes with other chemokines (Figure 4B). Other CXC family chemokine genes like CXCL1, CXCL2, CXCL3, and CXCL8 were not induced by LIGHT (Figure 4B). CXCL10 was not expressed in unstimulated HLF, and LIGHT also did not induce transcription of this chemokine (data not shown).

Furthermore, we found that LIGHT promoted increased gene expression of innate cytokines that are implicated in activation of ILC2s as well as other cell types, with transcription of IL-33 being strongly induced and TSLP elevated to a lesser extent (Figure 4C). Gene expression of the growth/differentiation factors IL-6 and GM-CSF was additionally enhanced by LIGHT. In contrast, genes expression of VEGF, TGF-β, and Activin A, that have also been implicated in lung inflammatory and remodeling responses, were unaffected (Figure 4C). Moreover, we found a selective activity of LIGHT in enhancing transcription of MMP-9 and ADAM8, two molecules implicated in modulating extracellular matrix deposition in the lungs. No effect was seen on the metalloproteinase genes, MMP-2 and -13, or the tissue inhibitor of metalloproteinases TIMP-1 (Figure 4D). Interestingly, LIGHT-induced gene expression was again significantly decreased with siRNA to LTβR, but not siRNA to HVEM (Figure 4E and not shown). As with proliferation, maximal activity of LIGHT in promoting the above factors was generally seen at doses of 25–100 ng/ml (Figure S2 in Supplementary Material). Assessing the kinetics of gene transcription showed that LIGHT activity was evident within 24 h and either maintained or increased over 72 h, and that differential kinetics was unlikely to explain why LIGHT did not promote expression of certain genes (Figure S3 in Supplementary Material).

These results emphasize that LIGHT through LTβR can target a broad range of inflammatory genes in lung fibroblasts, although with some selectivity. We next analyzed whether LIGHT-induced gene transcription was sufficient to result in production of the relevant inflammatory proteins. Interestingly, only IL-6 and GM-CSF were detected in the supernatants of LIGHT-stimulated fibroblasts out of the primary genes that were upregulated (Figures 4F,G), and we could not detect levels of CCL5, CXCL11, TSLP, IL-33, ADAM8, or MMP-9 protein by ELISA (data not shown). The limit of detection of these assays varied from 3.1 to 7.8 pg/ml. It is possible we could not detect protein as it was made at very low levels below the detection limit, or a period of time longer than 3 days was required for these products to be secreted. However, as IL-6 and GM-CSF were found, and their gene transcription was stronger, we favor the notion that transcription of the other genes was not sufficient for promoting detectable levels of protein. This is supported by further experiments described below. These data suggest that although LIGHT can positively affect transcription of a number of inflammatory genes, acting alone it may primarily promote production of the inflammatory proteins ICAM-1, VCAM-1, IL-6, and GM-CSF.

Because of the restricted induction of inflammatory proteins by LIGHT, this suggested that additional cofactors might be required to enhance the transcription of the other LIGHT-targeted genes to levels sufficient for translation into significant amounts of protein. We previously found that LIGHT could promote IL-13 production from eosinophils, and that IL-13 played an important role in aspects of the lung remodeling response in vivo that was dependent on LIGHT (11, 12). As lung fibroblasts express functional IL-4/IL-13 receptors, and IL-13 has been reported to exert some activities that might overlap with those we found for LIGHT (5, 6), we asked if IL-13 could reproduce some of the LIGHT activities in HLF. We also tested if IL-13 could synergize with LIGHT to promote certain inflammatory mediators. We confirmed the expression of IL-4Ralpha and IL-13Ralpha1, the receptors for IL-13, in cultured HLF by flow cytometry and found no change in their levels after LIGHT stimulation (data not shown).

First, we assessed whether IL-13 could promote division of HLF similar to LIGHT, or have any effect on alpha smooth muscle actin expression. As shown in Figure 5A, soluble IL-13 alone moderately increased HLF division, although not to statistically significant levels over that seen with unstimulated cells. Interestingly, simultaneous stimulation of HLF cells with LIGHT and IL-13 significantly enhanced proliferation of HLF cells compared with that seen with LIGHT or IL-13 alone. No change in the expression of HVEM or LTβR was observed after IL-13 stimulation (data not shown). Thus, both LIGHT and IL-13 can influence lung fibroblast proliferation, either at independent times or together. Some reports have also suggested that IL-13 can enhance alpha smooth muscle actin expression in lung fibroblasts (18, 19), while others have not found such an activity (20). We did not detect any effect of IL-13 alone, or in combination with LIGHT, in promoting the induction of alpha smooth muscle actin (data not shown).

Enhancement of gene transcription for the adhesion molecules ICAM-1 and VCAM-1 was also a shared feature of IL-13. It had a lesser effect on ICAM-1 mRNA than LIGHT, but together the two molecules resulted in enhanced ICAM-1 transcript levels compared to either molecule alone. VCAM-1 mRNA was induced by IL-13 alone to a greater extent compared to LIGHT, but no further expression resulted after stimulation with both molecules (Figure 5B). Flow cytometry analysis of these proteins revealed the same pattern as mRNA with LIGHT in combination with IL-13 resulting in higher levels of ICAM-1 (Figure 5B) but not VCAM-1 (data not shown).

We found divergent effects on the other LIGHT-regulated genes when HLF were exposed to LIGHT and IL-13 simultaneously. Transcription of CCL5 and ADAM8 was upregulated to lesser or equivalent extents by IL-13 alone compared to LIGHT, but expression of both of these genes was further enhanced when HLF were stimulated with LIGHT together with IL-13, either in additive or synergistic manners (Figure 5C). However, interestingly this still did not result in detection of either protein in the fibroblast supernatants (data not shown), suggesting additional stimulants are required for production of these molecules or alternatively the time point assayed was not sufficiently late for their detection. As described earlier, LIGHT did not have any significant impact on mRNA expression of CCL2 and CCL26 that were induced in HLF by IL-13, and LIGHT when combined with IL-13 did not modulate these IL-13-driven activities (Figure 5D). Similar to LIGHT, IL-13 action was not sufficient to induce secretion of CCL26 to levels above the ELISA detection limit, and stimulation with IL-13 together with LIGHT did not induce detectable protein (not shown). Interestingly, transcription of several inflammatory molecule genes that were induced by LIGHT alone was not significantly affected by IL-13 alone, namely CCL20, CXCL5, CXCL11, CXCL12, TSLP, IL-33, and MMP-9. Moreover, IL-13 partially or fully antagonized the enhancing effect of LIGHT on these genes (Figure 5E).

Of the two detectable proteins in supernatants of LIGHT-stimulated cells, IL-6 was induced at higher levels by IL-13 compared to LIGHT alone, but no further secretion was found when LIGHT was added with IL-13 (Figure 5F). GM-CSF protein was detected in IL-13-stimulated cultures at levels similar to those induced by LIGHT, and IL-13 and LIGHT synergized together for production of this protein (Figure 5G).

Lastly, we asked whether the inability of LIGHT or IL-13 to result in production of detectable levels of protein from certain genes, while having an obvious effect on mRNA, might be related to the level of transcription of these genes in HLF. Previously we found that LIGHT could induce mRNA, and production of protein, in primary HBE cells of several of the same molecules we found were targets in HLF (15). We then compared HLF to HBE in their response to LIGHT and IL-13. Significantly, LIGHT induced ICAM-1 mRNA at equivalent levels in HLF and HBE (Figure 6A), correlating with our ability to detect upregulation of ICAM-1 protein in both cell types [see above and Ref. (15)]. However, transcription of CCL5 and MMP-9 induced by LIGHT was markedly higher in HBE compared to HLF. The same pattern of synergy and antagonism, respectively, was observed in both cells when LIGHT was cultured in the presence of IL-13 (Figure 6A). Correlating with this, CCL5 and MMP-9 protein were readily detectable in HBE stimulated with LIGHT, and CCL5 production was further elevated while MMP-9 production was suppressed in the presence of IL-13 (Figures 6B,C). Thus, the action of LIGHT and IL-13 in modulating mRNA for certain genes is highly reproducible and significant, and shared between lung fibroblasts and lung epithelial cells, and the lack of production of significant levels of the protein products in fibroblasts is likely related to the overall level of gene transcription.