Get 20M+ Full-Text Papers For Less Than $1.50/day. Start a 7-Day Trial for You or Your Team.

Learn More →

Differential knockdown of TGF-β ligands in a three-dimensional co-culture tumor- stromal interaction model of lung cancer

Differential knockdown of TGF-β ligands in a three-dimensional co-culture tumor- stromal... Background: Transforming growth factor (TGF)-β plays a pivotal role in cancer progression through regulating cancer cell proliferation, invasion, and remodeling of the tumor microenvironment. Cancer-associated fibroblasts (CAFs) are the predominant type of stromal cell, in which TGF-β signaling is activated. Among the strategies for TGF-β signaling inhibition, RNA interference (RNAi) targeting of TGF-β ligands is emerging as a promising tool. Although preclinical studies support the efficacy of this therapeutic strategy, its effect on the tumor microenvironment in vivo remains unknown. In addition, differential effects due to knockdown of various TGF-β ligand isoforms have not been examined. Therefore, an experimental model that recapitulates tumor–stromal interaction is required for validation of therapeutic agents. Methods: We have previously established a three-dimensional co-culture model of lung cancer, and demonstrated the functional role of co-cultured fibroblasts in enhancing cancer cell invasion and differentiation. Here, we employed this model to examine how knockdown of TGF-β ligands affects the behavior of different cell types. We developed lentivirus vectors carrying artificial microRNAs against human TGF-β1 and TGF-β2, andtestedtheir effectsinlung cancer cells and fibroblasts. Results: Lentiviral vectors potently and selectively suppressed the expression of TGF-β ligands, and showed anti-proliferative effects on these cells. Furthermore, knockdown of TGF-β ligands attenuated fibroblast-mediated collagen gel contraction, and diminished lung cancer cell invasion in three-dimensional co-culture. We also observed differential effects by targeting different TGF-β isoforms in lung cancer cells and fibroblasts. Conclusions: Our findings support the notion that RNAi-mediated targeting of TGF-β ligands may be beneficial for lung cancer treatment via its action on both cancer and stromal cells. This study further demonstrates the usefulness of this three-dimensional co-culture model to examine the effect of therapeutic agents on tumor–stromal interaction. Keywords: RNA interference, MicroRNA, Lentivirus vector, TGF-β, Three-dimensional co-culture, Gel contraction assay * Correspondence: [email protected] Department of Respiratory Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Division for Health Service Promotion, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Full list of author information is available at the end of the article © 2014 Horie et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Horie et al. BMC Cancer 2014, 14:580 Page 2 of 11 http://www.biomedcentral.com/1471-2407/14/580 Background ligands have successfully ameliorated outcomes in dis- Lung cancer causes the deaths of more than one million ease models [16], and raised hope that this approach people worldwide every year [1]. Despite recent progress may be useful in a clinical setting. in molecular-targeted therapeutics, such as inhibitors of However, the three isoforms of TGF-β ligands—TGF- epidermal growth factor receptor (EGFR) tyrosine kinase β1, TGF-β2, and TGF-β3—show different expression pro- and anaplastic lymphoma kinase (ALK), failure to achieve files in various tissues and cell types. To develop effective long-lasting therapeutic responses has emphasized the therapeutic strategies for silencing TGF-β ligands, iden- need for novel treatment strategies [2,3]. tifying the appropriate isoform and target cell type may Most forms of cancer are associated with a stromal be critical. To our knowledge, the differential effects of response and extracellular matrix (ECM) deposition, eliminating specific TGF-β isoforms in a given tissue referred to as desmoplasia, which is critically regulated type remain unstudied. by cancer-associated fibroblasts (CAFs) [4]. Cancer tissue In the present study, we explored the therapeutic remodeling allows tumor cells to grow and disseminate, effect of TGF-β signaling blockade in lung cancer. and contributes to increased interstitial fluid pressure, We previously developed a three-dimensional (3D) which can be an obstacle to drug delivery [5]. co-culture model for evaluation of tumor–stromal inter- Among the soluble factors involved in the tumor–stro- actions [17]. Using this model, we tested the differential mal interaction, transforming growth factor (TGF)-β plays effects of silencing TGF-β ligands in A549 lung cancer a pivotal role. In premalignant stages, TGF-β acts as a cells and HFL-1 lung fibroblasts. Among the three iso- tumor suppressor by inhibiting proliferation and apoptotic forms of TGF-β ligands, TGF-β1 and TGF-β2 (but not induction in epithelial cells. In later stages, epithelial cells TGF-β3) are dominantly expressed in these cells [18-20]. become refractory to the growth inhibitory effect of TGF- Thus we established lentiviral vectors that transduce artifi- β and begin to secrete high levels of TGF-β, which in turn cial miRNAs against human TGF-β1and TGF-β2 as a tool exhibits tumor-promoting activity, such as angiogenesis, for testing the effects of TGF-β ligand knockdown. immune evasion, fibroblast activation, and ECM accumu- lation [6-8]. Furthermore, TGF-β increases the migratory Methods and invasive capacity of cancer cells by inducing the Cell culture epithelial–mesenchymal transition (EMT) [9,10]. Indeed, Tissue culture media and supplements were purchased TGF-β levels in both serum and tissues were elevated and from GIBCO (Life Technologies, Grand Island, NY). associated with worsening prognosis in patients with lung A549 human lung adenocarcinoma cells and HFL-1 cancer [11,12]. As such, TGF-β may be a promising target human lung fibroblasts were purchased from the American for cancer therapy. However, in contrast to cancer cells, Type Culture Collection (Rockville, MD), and were cul- the role of TGF-β signaling in the tumor stroma is poorly tured in Dulbecco’s Modified Eagle’sMedium(DMEM) understood, at least partly due to technical limitations in supplemented with 10% fetal bovine serum (FBS). In detecting TGF-β signaling activation in situ. addition, 293FT cells were obtained from Invitrogen RNA interference (RNAi) has been used widely to in- (Carlsbad, CA), and cultured in 100-mm dish coated duce the potent and specific inhibition of gene expres- with collagen type I (IWAKI, Tokyo, Japan) in DMEM sion. Several variants of small regulatory RNAs are with 10% FBS and 1 mM sodium pyruvate. involved in RNAi, including synthetic double-stranded small interfering RNAs (siRNAs), RNA polymerase III Artificial miRNA sequences (pol III)-transcribed small hairpin RNAs (shRNAs), and The BLOCK-iT™ Pol II miR RNAi Expression Vector Kit endogenous or artificial microRNAs (miRNAs) that are with EmGFP (Invitrogen, Carlsbad, CA) was used for transcribed by RNA polymerase II (pol II) as pri-miRNA, RNAi experiments. The design of the expression vector and subsequently processed into mature miRNAs [13,14]. was based on the use of endogenous murine miR-155 Vectors that enable the expression of engineered miRNA flanking sequences. Artificial miRNA sequences target- sequences from Pol II promoters have been developed ing human TGF-β ligands were designed using BLOCK- [15], in which the stem sequences of an endogenous iT™ RNAi Designer (http://rnaidesigner.Invitrogen.com/ miRNA precursor are substituted with unrelated base- rnaiexpress/). Four and three pairs of sense and anti- paired sequences that target specific genes. sense oligonucleotides were designed for targeting hu- Among the therapeutic strategies for TGF-β signaling man TGF-β1 and β2, respectively (Additional file 1: inhibition, RNAi is emerging as a promising tool [13]. Table S1). Recent advances in RNAi technology are overcoming previous obstacles, such as instability in vivo, impeded Plasmid construction and preparation of viral vectors drug delivery, and undesirable off-target effects. In ani- The designed oligonucleotides were annealed, followed mal experiments, RNAi agents directed against TGF-β by ligation into the pcDNA6.2-GW/EmGFP-miR vector Horie et al. BMC Cancer 2014, 14:580 Page 3 of 11 http://www.biomedcentral.com/1471-2407/14/580 (Invitrogen), which facilitates transfer into a suitable des- The optical density of each reaction was measured at tination vector via Gateway recombination reactions. 450 nm using a microplate reader (Bio-Rad, Hercules, CA), The EmGFP forward sequence primer (5′- GGCATG- and corrected against absorption at 570 nm. The GACGAGCTGTACAA −3′) was used for sequencing of data were analyzed using the Microplate Manager III the miRNA insert fragments, which was performed Macintosh data analysis software (Bio-Rad). using an ABI PRISM® 310 Genetic Analyzer. As the con- trol, pcDNA6.2-GW/EmGFP-miR negative control plas- Cell proliferation assay mid (Invitrogen) was used. The sequence containing the A549 cells were seeded at a density of 1 × 10 /well on miRNA coding region was transferred to the lentivirus 12-well dishes and HFL-1 cells were seeded at 4 × 10 / vector via the Gateway cloning system (Invitrogen). well on 6-well dishes. Both cell types were cultured in Briefly, the miRNA coding region was subcloned into DMEM containing 10% FBS. Cells were counted on days the entry plasmid pDONR221 (Invitrogen) using Gateway® 1, 3, and 5 after seeding using a hemocytometer. BP Clonase™ II Enzyme Mix (Invitrogen). The sequences in the entry plasmids were then transferred to the lenti- Collagen gel contraction assay and 3D co-culture viral expression vector, pCSII-EF-RfA, using Gateway® LR Three-dimensional gel cultures were carried out accord- Clonase™ II Enzyme Mix (Invitrogen). ing to the previously published protocol [17]. Briefly, collagen gels were prepared by mixing 0.5 mL of fibro- Lentivirus infection blast cell suspension (~2.5 × 10 cells) in FBS, 2.3 mL of The recombinant lentivirus was produced by transfec- type I collagen (Cell matrix type IA; Nitta Gelatin, tion of 293FT cells with the lentiviral expression vec- Tokyo, Japan), 670 μLof5×DMEM, and330 μLofre- tors, pCMV-VSV-G-RSV-Rev, and pCAG-HIVgp, using constitution buffer, following the manufacturer’srec- Lipofectamine 2000 reagent (Invitrogen). After 72 h, the ommendations. The mixture (3 mL) was cast into each medium was collected, and 1 × 10 of A549 or HFL-1 well of the six-well culture plates. The solution was cells were infected with 500 μL of medium containing then allowed to polymerize at 37°C for 30 min. After lentiviruses. For double knockdown of TGF-β1and overnight incubation, each gel was detached and cultured TGF-β2, 250 μL of each lentivirus-containing medium in growth medium, and the surface area of the gels was were used. Infection efficiency was assessed by measur- quantified via densitometry (Densitograph, ATTO, Tokyo, ing the percentage of EmGFP-positive cells via flow cy- Japan) for 5 consecutive days, and the final size relative to tometry (EPICS XL System II; Beckman Coulter, Brea, initial size was determined. For 3D co-culture, A549 cells CA), and knockdown efficiency of target gene was ana- (2 × 10 ) were seeded on the surface of each gel prior to lyzed using an enzyme-linked immunosorbent assay overnight incubation. After 5 days of floating culture, the (ELISA). gel was fixed in formalin solution and embedded in paraf- fin, and vertical sections were stained with hematoxylin RT-PCR and eosin. Total RNA was extracted using the RNeasy Mini Kit (Qiagen, Tokyo, Japan). The cDNA was synthesized Statistics using SuperScript III Reverse Transcriptase (Invitrogen), Results were confirmed by performing experiments in following the manufacturer’s protocol. Quantitative triplicate. Analyses were performed using JMP version reverse transcription (RT)-PCR was performed using 9 (SAS Institute Inc., Tokyo, Japan). For statistical sig- Mx-3000P (Stratagene, La Jolla, CA) and QuantiTect nificance, differences between two experimental groups SYBR Green PCR (Qiagen). Relative mRNA expression were examined using Student’s t-test, and Dunnett’stest was calculated using the ΔΔC method, and expression was used for multiple comparisons with control group. was normalized to that of the glyceraldehyde 3-phosphate P< 0.05 was considered to indicate significance. dehydrogenase (GAPDH) gene. The specific primers are shown in Additional file 2: Table S2. GSEA (gene set enrichment analysis) Navab et al. reported gene expression profiles for 15 pairs ELISA for TGF-β1 and TGF-β2 of lung CAFs and NFs, and identified genes enriched in A549 and HFL-1 cells were serum-starved for 24 h, and lung CAFs [21]. GSEA was performed using these micro- each supernatant was collected. The concentrations of array data sets (GSE22862) deposited in the public data- TGF-β1and TGF-β2 were measured using the Quanti- base. To obtain a gene set regulated by TGF-β,we used kine ELISA for human TGF-β1/TGF-β2(R&DSystems, publicly available microarray datasets, derived from two Minneapolis, MN), according to the manufacturer’sin- lung fibroblast cell lines stimulated by TGF-β:HFL-1 structions. Each supernatant was activated by 1 N HCl, (GSE27597) and IMR-90 (GSE17518) [22,23]. We ex- followed by neutralization with 1.2 N NaOH/0.5 M HEPES. tracted the top 800 TGF-β-induced genes from each Horie et al. BMC Cancer 2014, 14:580 Page 4 of 11 http://www.biomedcentral.com/1471-2407/14/580 dataset, as identified through the Significance Analysis enriched in CAFs, suggesting that TGF-β signaling is ac- of Microarrays (SAM) method. Combining these two tivated in lung CAFs (Figure 1A). We further extracted gene lists, we isolated 196 commonly induced genes in 88 ‘leading edge genes’ out of the TGF-β-regulated two lung fibroblast cell lines, which were defined as genes. A heatmap of these leading edge genes clearly ‘TGF-β-regulated genes’ (Additional file 3: Table S3). illustrated differential expression between CAFs and NFs (Figure 1B). As expected, ECM-related genes were Results enriched among the leading edge genes, and a heatmap of TGF-β signaling is activated in lung CAFs 16 selected ECM related genes apparently showed that CAFs are a major constituent of the tumor stroma, and TGF-β-regulated ECM-related enzymes and substrates, we have previously shown that lung CAFs are more potent including PLOD1, LOX,COL1A1, VCAN,SPARC,FN1, in enhancing cancer cell invasion and collagen gel con- ELN,and THBS1, aremoreenrichedinCAFsthanNFs traction than normal lung fibroblasts (NFs) [17]. Although (Figure 1C). theroleofTGF-β in cancer cells and lung fibroblasts has been investigated extensively, TGF-β function in CAFs Lentivirus-mediated transduction of artificial miRNAs remains largely unknown due to technical hurdles in against human TGF-β1 and TGF-β2 isolating fibroblasts from lung cancer tissues. Based on the observation that endogenous TGF-β signal- To examine TGF-β signaling activation status in lung ing is activated in lung CAFs, we examined whether TGF- CAFs, we used gene set enrichment analysis (GSEA) to β signaling activation in fibroblasts modulates the behavior determine whether the expression of the identified TGF- of adjacent cancer cells. We also aimed to elucidate the β-regulated genes was enhanced in lung CAFs compared cell-autonomous action of TGF-β in lung cancer cells. To to NFs. This was performed using microarray data sets this end, we generated lentiviral vectors that transduced of CAFs and NFs reported by Navab et al. [21]. These artificial miRNAs against TGF-β ligands, and tested their analyses demonstrated that the TGF-β-regulated genes effects on lung cancer cells and HFL-1 lung fibroblasts. identified through our analysis are in fact highly The expression levels of TGF-β isoforms are variable Figure 1 Gene set enrichment analysis (GSEA). A: GSEA was used to examine the enrichment of identified TGF-β-regulated genes in CAFs. ‘TGF-β-regulated genes’ include 196 genes induced by TGF-β in both IMR-90 and HFL-1 lung fibroblast cell lines. CAF and NF gene expression profiles reported by Navab et al. [21] were used. Enrichment of TGF-β-regulated genes is shown schematically with those that best correlated with the CAF phenotype on the left (‘CAF-high’) and the genes that best correlated with the NF phenotype on the right (‘NF-high’). B: A heat map representing the relative expression change of ‘ 88 leading edge genes’ which were obtained by GSEA analysis in CAFs and NFs. C: A heat map representing the relative expression change of selected ‘16 ECM related genes’. Horie et al. BMC Cancer 2014, 14:580 Page 5 of 11 http://www.biomedcentral.com/1471-2407/14/580 among lung cancer cell lines. In order to survey these dif- against TGF-β1 were ineffective for TGF-β2 (Figure 2B, ferences, we used Cancer Cell Line Encyclopedia (CCLE) right). These results show that miRNAs against TGF-β1 data and found that expression of TGF-β isoforms are or TGF-β2 exert their effects in a selective manner for relatively high in A549 cells among 111 non-small cell each ligand. Out of the two combinations tested for lung cancer cell lines (Additional file 4: Figure S1). There- double knockdown, miRNA #2 against TGF-β1and #2 fore, we used A549 lung cancer cells in the following against TGF-β2 showed efficient silencing in both A549 experiments. and HFL-1 cells (Figure 2). Therefore, we selected miRNA Four miRNA sequences were designed to target hu- sequences #2 against TGF-β1 and #2 against TGF-β2, for man TGF-β1, as well as three sequences against TGF-β2 single or double knockdown in the following experiments. (Additional file 1: Table S1). Next, we determined the ef- ficiency of lentiviral infection by measuring the percent- Cell proliferation is suppressed by knockdown of TGF-β1 age of EmGFP-positive cells using flow cytometry. More and/or TGF-β2 than 95% of A549 cells were positive for EmGFP, sug- Next, we investigated whether TGF-β1 and/or TGF-β2 gesting a high transduction efficiency for this miRNA knockdown affected the proliferation of A549 and HFL-1 sequence (Additional file 5: Figure S2A, left); we ob- cells. In both cell types, the transduction of artificial served similar efficiencies for all miRNA sequences used miRNAs against TGF-β1or TGF-β2 suppressed cell in this study (Additional file 5: Figure S2A, right). proliferation (Figure 3), and this anti-proliferative effect Meanwhile, HFL-1 cells showed more modest (but still was enhanced in cells subject to double knockdown, com- sufficient) efficiencies for lentiviral infection (Additional pared to single knockdown of either TGF-β1orTGF-β2. file 5: Figure S2B, left). The percentage of EmGFP- TGF-β is a strong inhibitor of proliferation in most positive cells ranged from 65–85% among the miRNA epithelial cells, whereas it promotes proliferation in mes- sequences (Additional file 5: Figure S2B, right). enchymal cells and enhances cancer cell survival [6-8]. For double knockdown of TGF-β1 and TGF-β2, two Our lentivirus-mediated miRNA delivery system main- combinations of lentiviruses encoding miRNAs against tains stable knockdown of TGF-β1 and/or TGF-β2. This TGF-β1and TGF-β2 were co-infected: #2 miRNA against may alter cell signaling in the steady state and modulate TGF-β1and #2 miRNA against TGF-β2 (TGF-β1KD #2+ the cell machinery that regulates cell survival or prolifer- TGF-β2KD #2), or #4 miRNA against TGF-β1and #3 ation, thereby resulting in suppressed cell proliferation. miRNA against TGF-β2(TGF-β1KD #4+ TGF-β2KD #3). Co-infection with two different lentiviruses showed similar transduction efficiencies compared to single infections, as Altered EMT-related gene expression via TGF-β1 and/or determined via EmGFP fluorescence (Additional file 5: TGF-β2 knockdown Figure S2A, right and Additional file 5: Figure S2B, right). EMT is crucial for cancer cells to acquire invasive pheno- types, which are characterized by downregulation of E- Potent and selective knockdown of TGF-β1 and TGF-β2 cadherin and upregulation of vimentin. A549 cells stay in Next, we evaluated the efficiency of TGF-β knockdown an intermediary state of EMT, whereas exogenous TGF-β through measurement of protein expression via ELISA. further promotes acquisition of mesenchymal phenotypes To control for unintended effects of experimental ma- [20]. We examined whether knockdown of TGF-β ligands nipulation, we examined the expression of TGF-β1 and modulated the expression of EMT markers. TGF-β2 in uninfected A549 and HFL-1 cells compared Silencing of TGF-β2 led to E-cadherin upregulation, to cells infected with negative control (NTC) miRNAs suggesting the restoration of epithelial phenotypes. In ac- (Figure 2). No significant difference in TGF-β1orTGF-β2 cordance, vimentin expression was suppressed by knock- expression was observed. down of TGF-β1and/or TGF-β2, though it failed to reach In A549 cells, three of four miRNAs against TGF-β1 statistical significance (Figure 4A). These results support (#1, #2, and #4) were able to silence TGF-β1 expression, the notion that endogenous TGF-β signaling participates whereas all three miRNAs against TGF-β2 were ineffective in the maintenance of a mesenchymal phenotype in A549 for TGF-β1 (Figure 2A, left). Two out of three miRNAs cells in the steady state. against TGF-β2 (#2 and #3) silenced TGF-β2 expression, EMT is accompanied by the enhanced expression whereas all four miRNAs against TGF-β1 were ineffective of fibrogenic growth factors, such as platelet-derived for TGF-β2 (Figure 2A, right). In HFL-1 cells, three of growth factor (PDGF) and connective tissue growth fac- four miRNAs against TGF-β1 (#1, #2 and #3) were able to tor (CTGF) [20]. PDGF is a dimeric protein composed silence TGF-β1 expression, whereas all three miRNAs of A and B subunits, and it has been reported that the against TGF-β2 were ineffective for TGF-β1 (Figure 2B, transcription of PDGFB is regulated by TGF-β.Consist- left). Two of three miRNAs against TGF-β2 (#2 and #3) si- ent with the previous experiment [20], TGF-β2 silencing lenced TGF-β2 expression, whereas all four miRNAs led to CTGF downregulation, whereas knockdown of Horie et al. BMC Cancer 2014, 14:580 Page 6 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 2 Knockdown of TGF-β ligands. A: TGF-β1 and TGF-β2 concentrations measured by ELISA in the supernatant of A549 cells transduced with each miRNA. Left: TGF-β1. Right: TGF-β2. Data shown are the means ± SEM of triplicate analyses. KD: knockdown. NTC: negative control. The concentration of TGF-β1 or TGF-β2 in the supernatant of cells with TGF-β1 and/or TGF-β2 knockdown was compared to that of cells transduced with NTC miRNA. Statistical significance was determined by Dunnett’s test. P< 0.05. B: TGF-β1 and TGF-β2 concentrations measured by ELISA in the supernatant of HFL-1 cells. TGF-β1 and/or TGF-β2 attenuated PDGFB expression TGF-β1 and/or TGF-β2 knockdown attenuates collagen (Figure 4B). gel contraction in HFL-1 cells Upon TGF-β stimulation, fibroblasts convert to an acti- Cancer tissue contraction facilitates tumor progression vated phenotype to enhance ECM production. Thus, we and contributes to increased interstitial fluid pressure, examined whether knockdown of TGF-β1and/orTGF-β2 which hampers drug delivery [5]. The collagen gel con- modulated the expression of α1 (I) collagen (COL1A1), a traction assay is used widely to recreate tissue contraction major component of ECM. In HFL-1 cells, TGF-β1 knock- in an experimental setting, and it has been shown that down decreased the expression of COL1A1, whereas TGF-β stimulates fibroblast-mediated collagen gel con- TGF-β2 silencing had no effect (Figure 4C). traction [25]. We used this assay to investigate whether These results suggest the differential regulation of knockdown of TGF-β1 and/or TGF-β2 modulated tissue target genes by TGF-β1orTGF-β2 in cancer cells and contraction through effects on fibroblasts. fibroblasts. During lung branching morphogenesis, Collagen gels were embedded with HFL-1 cells after TGF-β1 expression is prominent throughout the mes- TGF-β1 and/or TGF-β2 knockdown, and gel size was enchyme, whereas TGF-β2 is localized to mainly the measured daily. On the first day, the control gel size was epithelium of the developing distal airways [24]. Thus, reduced to ~50% of the initial value, followed by gradual TGF-β2 may be critical for determining epithelial or shrinkage to less than 20% on the fifth day (Figure 5). cancer cell behavior in a cell-autonomous fashion, Compared to the control, knockdown of TGF-β1and/or whereas endogenous TGF-β1 may play a greater role in TGF-β2 in HFL-1 cells attenuated gel contraction (Figure 5 fibroblasts. and Additional file 6: Figure S3). These results suggested Horie et al. BMC Cancer 2014, 14:580 Page 7 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 3 Cell proliferation assay. Cell proliferation curve in A549 or HFL-1 cells transduced with NTC miRNA (solid line) compared to cells transduced with miRNA against TGF-β1 (dashed line: TGF-β1KD),TGF-β2(dotted line: TGF-β2KD),orTGF-β1 and TGF-β2 (dashed-dotted line: TGF-β1+ β2KD).Cell counts were carried out on days 1, 3, and 5 after seeding. Left: A549. Right: HFL-1. Data shown are the means ± SEM of triplicate analyses. Numbers of cells with TGF-β1and/orTGF-β2 knockdown on day 5 was compared to that in the cells transduced with NTC miRNA. Statistical significance was determined by Student’s t-test. P< 0.05. that the inhibition of endogenous TGF-β signaling in Discussion fibroblasts ameliorates tissue contraction. TGF-β plays several crucial roles in cancer progression, affecting both tumor and stromal cells, including fi- broblasts [4]. However, very little is known regarding Three-dimensional co-culture of A549 and HFL-1 cells the effects of TGF-β ligand silencing in the context To examine the interaction between lung cancer cells of tumor–stromal or epithelial–mesenchymal interac- and fibroblasts, we previously established a 3D co-culture tions [26]. Numerous reports have shown the effects model [17]. HFL-1 cells transduced with control miRNAs of exogenous TGF-β stimulation in various cell types, or those for TGF-β1and TGF-β2 silencing (double knock- whereas the effects of endogenous or cell-autonomous down) were embedded into the collagen gels, and then TGF-β signaling are poorly understood. To our know- A549 cells were seeded onto the surface of these gels. The ledge, this study is the first to generate lentiviral vectors co-cultured collagen gels were subjected to floating cul- encoding artificial miRNAs targeting human TGF-β1and ture for an additional 5 days, followed by hematoxylin and TGF-β2, and to explore their effects in a co-culture eosin staining (Figure 6). model. Double knockdown of TGF-β1 and TGF-β2 in HFL-1 Lentiviral vectors showed efficient transduction in A549 cells did not show clear effects on A549 cell invasion, lung cancer cells, as well as HFL-1 lung fibroblasts. suggesting a minor role for TGF-β produced in HFL-1 Knockdown efficiency to less than 30% of the control was cells in this co-culture model (lower panels). In our obtained for both TGF-β1and TGF-β2 in a selective previous work, we did not examine whether HFL-1 manner. Knockdown of TGF-β ligands suppressed cell cells enhance lung cancer cell invasion [17], and this proliferation in both A549 and HFL-1 cells. Furthermore, study suggests that endogenous TGF-β expression in expression of EMT markers and fibrogenic growth factors HFL-1 cells may not have a significant role in invasion was modulated in A549 cells, whereas collagen I was promotion. downregulated in HFL-1 cells. With regard to cellular In contrast, A549 cell invasion was observed when function, silencing of TGF-β ligands attenuated HFL-1- control A549 cells were cultured with control HFL-1 mediated collagen gel contraction, and inhibited A549 cell cells (upper left panel). Silencing of either TGF-β1or invasion in the 3D co-culture model. All of these findings TGF-β2 in A549 cells failed to inhibit invasion (upper support the tumor-promoting role of TGF-β,and that middle panels), whereas double knockdown of TGF-β1 the reported beneficial effects of TGF-β inhibition in and TGF-β2 led to complete disappearance of invading cancer therapeutics may derive from interfering with cells (upper right panel). tumor–stromal communications. Horie et al. BMC Cancer 2014, 14:580 Page 8 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 4 Quantitative RT-PCR. A: Quantitative RT-PCR for E-cadherin (left) and vimentin (right) in A549 cells. B: Quantitative RT-PCR for CTGF (left) and PDGFB (right) in A549 cells. C: Quantitative RT-PCR for COL1A1 in HFL-1 cells. Data shown are the means ± SEM. The relative expression of each gene in cells with TGF-β1 and/or TGF-β2 knockdown was compared to that in the cells transduced with NTC miRNA. Statistical significance was determined by Student’s t-test. P< 0.05. In our experiments, it appeared that both TGF-β1and and invasion in a 3D co-culture. In HFL-1 cells, TGF-β1 TGF-β2wereabundantlyproducedinA549 cells,whereas knockdown was more effective than TGF-β2knockdown the concentration of TGF-β1 was higher than that of in suppressing COL1A1 expression. TGF-β2 in the supernatant of HFL-1 cells. Compared to Little is known regarding the expression profiles of single knockdown, double knockdown of TGF-β1and TGF-β isoforms in various lung cancer cell types. As TGF-β2 showed stronger effects in A549 cell proliferation shown here, knockdown of each TGF-β ligand Horie et al. BMC Cancer 2014, 14:580 Page 9 of 11 http://www.biomedcentral.com/1471-2407/14/580 knockdown of TGF-β1 and TGF-β2 in HFL-1 cells did not show clear effects on A549 cell invasion, and en- NTC miRNA dogenous TGF-β expression in HFL-1 cells seemed to TGF- 1 KD have little effect on lung cancer cell invasion. The pre- TGF- 2 KD cise mechanism underlying CAF-enhanced lung cancer TGF- 1+ 2KD cell invasion remains to be elucidated, and further stud- ies are necessary to clarify the mechanisms underlying cell invasion in our experimental model. There have been several attempts to exploit TGF-β signaling inhibition as a therapeutic approach for malig- nant tumors, including the use of TGF-β receptor kinase inhibitors, TGF-β neutralizing antibodies, TGF-β anti- sense oligonucleotides (AONs), and siRNAs [27]. TGF-β type I receptor kinase inhibitor has been tested for non- small cell lung cancer (NSCLC) patients in a phase II study, but failed to yield clinical benefits [28]. Several animal models of cancer have demonstrated the thera- * peutic effect of TGF-β neutralizing antibodies [29]. Recently, AONs against TGF-β ligands have shown promising clinical results. Trabedersen (AP 12009) is an AON against human TGF-β2. Intra-tumoral administra- tion of trabedersen in patients with high-grade gliomas 01 2 3 4 5 led to better tumor control and prolonged survival with Day fewer adverse events, which prompted a larger phase III Figure 5 Collagen gel contraction assay. Time-course of gel trial [30]. Intravenous application of trabedersen in pa- contraction in the presence of HFL-1 transduced with NTC miRNA tients with other cancer types is also under evaluation. AP (solid line), or miRNAs against TGF-β1 (dashed line: TGF-β1KD), TGF-β2 11014, another AON targeting human TGF-β1, is cur- (dotted line: TGF-β2KD), orTGF-β1 and TGF-β2 (dashed-dotted line: rently in preclinical development for NSCLC treatment. TGF-β1+ β2 KD). The area of each gel was assessed daily for 5 days Furthermore, a phase II trial for belagenpumatucel-L, a and the relative value compared to the initial size was determined. Data shown are the means ± SD of triplicate analyses. Statistical vaccine produced from NSCLC cells transfected with significance was determined by Student’st-test. *P< 0.05. TGF-β2 AON, has shown beneficial effects on survival without any significant adverse effects; phase III studies in modulated phenotype in a cell-type-dependent manner. lung cancer patients are ongoing [31]. RNAi targeting These effects may be much more complicated and variable TGF-β ligands is also emerging as a promising tool [13]. depending on the multicellular context; nevertheless, our In animal experiments, RNAi agents against TGF-β1 results demonstrate the important role for TGF-β signal- demonstrated therapeutically beneficial effects, support- ing in the tumor microenvironment. ing progression toward future clinical applications [16]. We have reported previously that lung CAFs enhance This body of work demonstrates the intensifying interest cancer cell invasion [17]. In the present study, double in TGF-β ligand silencing as a therapeutic approach for Figure 6 3D co-culture model. Hematoxylin and eosin staining of 3D cultured gels composed of A549 and HFL-1 cells transduced with the indicated miRNAs. Upper panels: HFL-1 cells transduced with NTC miRNA. Lower panels: HFL-1 cells transduced with miRNAs against TGF-β1 and TGF-β2 (TGF-β1+ β2 KD). Invading cells are indicated with arrows. Scale bar: 100 μm. % of initial size of gel Horie et al. BMC Cancer 2014, 14:580 Page 10 of 11 http://www.biomedcentral.com/1471-2407/14/580 lung cancer. To validate therapeutic strategies against Authors’ contributions MH carried out the experiments and drafted the manuscript. AS, TK, and TN TGF-β ligands, it may be critical to target the appropriate designed the study and participated in manuscript preparation. SN and HIS TGF-β isoform in a given cell type. The present study pro- performed statistical analyses. MO participated in the design of the study. vides a useful experimental model to investigate the effect YM participated in preparation of tissue sections. All authors read and approved the final manuscript. of therapeutic agents targeting TGF-β ligands. Our results suggest that targeting both TGF-β1and TGF-β2inlung Acknowledgements cancer cells is more effective than single knockdown. Fur- This work was supported by KAKENHI (Grants-in-Aid for Scientific Research) thermore, TGF-β2 knockdown may play a more specific from the Ministry of Education, Culture, Sports, Science, and Technology, and a grant to the Respiratory Failure Research Group from the Ministry of role in lung cancer cells than in stromal cells, such as fi- Health, Labour and Welfare, Japan. We thank Makiko Sakamoto for the broblasts. Future studies are warranted to further elucidate technical assistance. the therapeutical benefits of strategies against the different Author details TGF-β ligands. Department of Respiratory Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Division for Health Service Promotion, The University of Tokyo, 7-3-1 Hongo, Conclusion Bunkyo-ku, Tokyo 113-0033, Japan. Department of Biochemistry, Nihon Because TGF-β exerts it pleiotropic effects in a variety University School of Dentistry, 1-8-13 Kanda-Surugadai, Chiyoda-ku, Tokyo of cells in the tumor microenvironment, it is useful to 101-8310, Japan. Department of Biochemistry, Ohu University School of Pharmaceutical Sciences, Misumido 31-1, Tomitamachi, Koriyama, Fukushima evaluate the action of anti-TGF-β therapeutic agents in 963-8611, Japan. Department of Molecular Pathology, Graduate School of multicellular culture conditions. Our 3D co-culture Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, model, demonstrated here, represents a useful tool for Japan. The Fourth Department of Internal Medicine, Teikyo University School of Medicine University Hospital, Mizonokuchi, 3-8-3 Mizonokuchi, evaluating differential effects on cancer cells and fibro- Takatsu-ku, Kawasaki, Kanagawa 213-8507, Japan. blasts. In summary, we established a lentivirus-mediated knockdown system for TGF-β ligands, which revealed Received: 8 February 2014 Accepted: 4 August 2014 Published: 9 August 2014 their multifaceted effects on cell proliferation, EMT, inva- sion, and ECM remodeling. References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D: Global cancer statistics. CA Cancer J Clin 2011, 61(2):69–90. Additional files 2. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H, Gemma A, Harada M, Yoshizawa H, Kinoshita I, Fujita Y, Okinaga S, Hirano H, Additional file 1: Table S1. Sequences of artificial miRNAs against Yoshimori K, Harada T, Ogura T, Ando M, Miyazawa H, Tanaka T, Saijo Y, TGF-β ligands. Hagiwara K, Morita S, Nukiwa T: Gefitinib or chemotherapy for non-small- cell lung cancer with mutated EGFR. N Engl J Med 2010, Additional file 2: Table S2. Primers for RT-PCR. 362(25):2380–2388. Additional file 3: Table S3. The 196 ‘TGF-β-regulated genes’. 3. Choi YL, Soda M, Yamashita Y, Ueno T, Takashima J, Nakajima T, Yatabe Y, Additional file 4: Figure S1. Expression levels of TGF-β isoforms in Takeuchi K, Hamada T, Haruta H, Ishikawa Y, Kimura H, Mitsudomi T, Tanio non-small cell lung cancer cell lines. The transcription levels of TGF-β1 Y, Mano H: EML4-ALK mutations in lung cancer that confer resistance to and TGF-β2 in non-small cell lung cancer cell lines were retrieved from ALK inhibitors. N Engl J Med 2010, 363(18):1734–1739. Cancer Cell Line Encyclopedia (CCLE) database and shown in a scatter 4. Micke P, Ostman A: Tumour-stroma interaction: cancer-associated plot. A549 cells showed relatively higher levels of TGF-β1 and TGF-β2. fibroblasts as novel targets in anti-cancer therapy? Lung Cancer 2004, 45(Suppl 2):S163–S175. Additional file 5: Figure S2. Transduction efficiency of lentiviral 5. Heldin CH, Rubin K, Pietras K, Ostman A: High interstitial fluid pressure - vectors. A: Transduction efficiency of miRNAs in A549 cells. Left: miRNA an obstacle in cancer therapy. Nat Rev Cancer 2004, 4(10):806–813. transduction was tracked by detecting EmGFP-positive cells using the 6. Massagué J: TGFbeta in Cancer. Cell 2008, 134(2):215–230. FL-1 channel of a flow cytometer. A representative result of #2 miRNA 7. Bierie B, Moses HL: Tumour microenvironment: TGFbeta: the molecular transduction against TGF-β1 is shown. The grey and black peaks are from Jekyll and Hyde of cancer. Nat Rev Cancer 2006, 6(7):506–520. uninfected and lentivirus-transduced cells, respectively. Right: transduction 8. Ikushima H, Miyazono K: TGFbeta signalling: a complex web in cancer efficiency of each miRNA. KD: knockdown. NTC: negative control. B: progression. Nat Rev Cancer 2010, 10(6):415–424. Transduction efficiency of miRNA in HFL-1 cells. 9. Saito A, Suzuki HI, Horie M, Ohshima M, Morishita Y, Abiko Y, Nagase T: An Additional file 6: Figure S3. Collagen gel contraction assay. integrated expression profiling reveals target genes of TGF-beta and Photographs of the gels on day 5 in the experiments shown in Figure 5. TNF-alpha possibly mediated by microRNAs in lung cancer cells. PLoS Identically sized white circles in each well are shown to demonstrate the One 2013, 8(2):e56587. differences in gel size. 10. Kalluri R, Weinberg RA: The basics of epithelial-mesenchymal transition. J Clin Invest 2009, 119(6):1420–1428. 11. Kong F, Jirtle RL, Huang DH, Clough RW, Anscher MS: Plasma transforming Abbreviations growth factor-beta1 level before radiotherapy correlates with long term TGF-β: Transforming growth factor-β; EGFR: Epidermal growth factor outcome of patients with lung carcinoma. Cancer 1999, 86(9):1712–1719. receptor; ALK: Anaplastic lymphoma kinase; CAFs: Cancer-associated 12. Hasegawa Y, Takanashi S, Kanehira Y, Tsushima T, Imai T, Okumura K: fibroblasts; EMT: Epithelial-mesenchymal transition; ECM: Extracellular matrix; Transforming growth factor-beta1 level correlates with angiogenesis, GSEA: Gene set enrichment analysis; PDGF: Platelet-derived growth factor; tumor progression, and prognosis in patients with nonsmall cell lung CTGF: Connective tissue growth factor; RNAi: RNA interference; carcinoma. Cancer 2001, 91(5):964–971. AON: Antisense oligonucleotides. 13. Davidson BL, McCray PB Jr: Current prospects for RNA interference-based therapies. Nat Rev Genet 2011, 12(5):329–340. Competing interests 14. Cullen BR: Transcription and processing of human microRNA precursors. The authors declare that they have no competing interests. Mol Cell 2004, 16(6):861–865. Horie et al. BMC Cancer 2014, 14:580 Page 11 of 11 http://www.biomedcentral.com/1471-2407/14/580 15. Zeng Y, Wagner EJ, Cullen BR: Both natural and designed micro RNAs can TGF-beta2 inhibitor trabedersen: results of a randomized and controlled inhibit the expression of cognate mRNAs when expressed in human phase IIb study. Neuro-Oncology 2011, 13(1):132–142. cells. Mol Cell 2002, 9(6):1327–1333. 31. Nemunaitis J, Dillman RO, Schwarzenberger PO, Senzer N, Cunningham C, 16. Hamasaki T, Suzuki H, Shirohzu H, Matsumoto T, D'Alessandro-Gabazza CN, Cutler J, Tong A, Kumar P, Pappen B, Hamilton C, DeVol E, Maples PB, Liu L, Gil-Bernabe P, Boveda-Ruiz D, Naito M, Kobayashi T, Toda M, Mizutani T, Chamberlin T, Shawler DL, Fakhrai H: Phase II study of belagenpumatucel-L, a Taguchi O, Morser J, Eguchi Y, Kuroda M, Ochiya T, Hayashi H, Gabazza EC, transforming growth factor beta-2 antisense gene-modified allogeneic Ohgi T: Efficacy of a novel class of RNA interference therapeutic agents. tumor cell vaccine in non-small-cell lung cancer. JClin Oncol 2006, PLoS One 2012, 7(8):e42655. 24(29):4721–4730. 17. Horie M, Saito A, Mikami Y, Ohshima M, Morishita Y, Nakajima J, Kohyama T, Nagase T: Characterization of human lung cancer-associated fibroblasts doi:10.1186/1471-2407-14-580 in three-dimensional in vitro co-culture model. Biochem Biophys Res Cite this article as: Horie et al.: Differential knockdown of TGF-β ligands Commun 2012, 423(1):158–163. in a three-dimensional co-culture tumor- stromal interaction model of lung cancer. BMC Cancer 2014 14:580. 18. Wen FQ, Kohyama T, Skold CM, Zhu YK, Liu X, Romberger DJ, Stoner J, Rennard SI: Glucocorticoids modulate TGF-beta production by human fetal lung fibroblasts. Inflammation 2003, 27(1):9–19. 19. Jakowlew SB, Mathias A, Chung P, Moody TW: Expression of transforming growth factor beta ligand and receptor messenger RNAs in lung cancer cell lines. Cell Growth Differ 1995, 6(4):465–476. 20. Saito RA, Watabe T, Horiguchi K, Kohyama T, Saitoh M, Nagase T, Miyazono K: Thyroid transcription factor-1 inhibits transforming growth factor- beta-mediated epithelial-to-mesenchymal transition in lung adenocarcinoma cells. Cancer Res 2009, 69(7):2783–2791. 21. Navab R, Strumpf D, Bandarchi B, Zhu CQ, Pintilie M, Ramnarine VR, Ibrahimov E, Radulovich N, Leung L, Barczyk M, Panchal D, To C, Yun JJ, Der S, Shepherd FA, Jurisica I, Tsao MS: Prognostic gene-expression signature of carcinoma-associated fibroblasts in non-small cell lung cancer. Proc Natl Acad Sci U S A 2011, 108(17):7160–7165. 22. Campbell JD, McDonough JE, Zeskind JE, Hackett TL, Pechkovsky DV, Brandsma CA, Suzuki M, Gosselink JV, Liu G, Alekseyev YO, Xiao J, Zhang X, Hayashi S, Cooper JD, Timens W, Postma DS, Knight DA, Marc LE, James HC, Avrum S: A gene expression signature of emphysema-related lung destruction and its reversal by the tripeptide GHK. Genome Med 2012, 4(8):67. 23. Hecker L, Vittal R, Jones T, Jagirdar R, Luckhardt TR, Horowitz JC, Pennathur S, Martinez FJ, Thannickal VJ: NADPH oxidase-4 mediates myofibroblast activation and fibrogenic responses to lung injury. Nat Med 2009, 15(9):1077–1081. 24. Bragg AD, Moses HL, Serra R: Signaling to the epithelium is not sufficient to mediate all of the effects of transforming growth factor beta and bone morphogenetic protein 4 on murine embryonic lung development. Mech Dev 2001, 109(1):13–26. 25. Saito RA, Micke P, Paulsson J, Augsten M, Peña C, Jönsson P, Botling J, Edlund K, Johansson L, Carlsson P, Jirström K, Miyazono K, Ostman A: Forkhead box F1 regulates tumor-promoting properties of cancer- associated fibroblasts in lung cancer. Cancer Res 2010, 70(7):2644–2654. 26. Kage H, Sugimoto K, Sano A, Kitagawa H, Nagase T, Ohishi N, Takai D: Suppression of transforming growth factor beta1 in lung alveolar epithelium-derived cells using adeno-associated virus type 2/5 vectors to carry short hairpin RNA. Exp Lung Res 2011, 37(3):175–185. 27. Hawinkels LJ, Ten Dijke P: Exploring anti-TGF-beta therapies in cancer and fibrosis. Growth Factors 2011, 29(4):140–152. 28. Scagliotti GV, Ilaria R Jr, Novello S, von Pawel J, Fischer JR, Ermisch S, de Alwis DP, Andrews J, Reck M, Crino L, Eschbach C, Manegold C: Tasisulam sodium (LY573636 sodium) as third-line treatment in patients with unresectable, metastatic non-small-cell lung cancer: a phase-II study. J Thorac Oncol 2012, 7(6):1053–1057. 29. Biswas S, Guix M, Rinehart C, Dugger TC, Chytil A, Moses HL, Freeman ML, Arteaga CL: Inhibition of TGF-beta with neutralizing antibodies prevents Submit your next manuscript to BioMed Central radiation-induced acceleration of metastatic cancer progression. J Clin and take full advantage of: Invest 2007, 117(5):1305–1313. 30. Bogdahn U, Hau P, Stockhammer G, Venkataramana NK, Mahapatra AK, Suri • Convenient online submission A, Balasubramaniam A, Nair S, Oliushine V, Parfenov V, Poverennova I, Zaaroor M, Jachimczak P, Ludwig S, Schmaus S, Heinrichs H, • Thorough peer review Schlingensiepen KH: Targeted therapy for high-grade glioma with the • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit http://www.deepdyve.com/assets/images/DeepDyve-Logo-lg.png BMC Cancer Springer Journals

Differential knockdown of TGF-β ligands in a three-dimensional co-culture tumor- stromal interaction model of lung cancer

Download PDF
11 pages

Loading next page...
 
/lp/springer-journals/differential-knockdown-of-tgf-ligands-in-a-three-dimensional-co-vGr0CDMYz0

References (39)

Publisher
Springer Journals
Copyright
Copyright © 2014 by Horie et al.; licensee BioMed Central Ltd.
Subject
Biomedicine; Cancer Research; Oncology; Surgical Oncology; Health Promotion and Disease Prevention; Biomedicine general; Medicine/Public Health, general
ISSN
1471-2407
eISSN
1471-2407
DOI
10.1186/1471-2407-14-580
pmid
25107280
Publisher site
See Article on Publisher Site

Abstract

Background: Transforming growth factor (TGF)-β plays a pivotal role in cancer progression through regulating cancer cell proliferation, invasion, and remodeling of the tumor microenvironment. Cancer-associated fibroblasts (CAFs) are the predominant type of stromal cell, in which TGF-β signaling is activated. Among the strategies for TGF-β signaling inhibition, RNA interference (RNAi) targeting of TGF-β ligands is emerging as a promising tool. Although preclinical studies support the efficacy of this therapeutic strategy, its effect on the tumor microenvironment in vivo remains unknown. In addition, differential effects due to knockdown of various TGF-β ligand isoforms have not been examined. Therefore, an experimental model that recapitulates tumor–stromal interaction is required for validation of therapeutic agents. Methods: We have previously established a three-dimensional co-culture model of lung cancer, and demonstrated the functional role of co-cultured fibroblasts in enhancing cancer cell invasion and differentiation. Here, we employed this model to examine how knockdown of TGF-β ligands affects the behavior of different cell types. We developed lentivirus vectors carrying artificial microRNAs against human TGF-β1 and TGF-β2, andtestedtheir effectsinlung cancer cells and fibroblasts. Results: Lentiviral vectors potently and selectively suppressed the expression of TGF-β ligands, and showed anti-proliferative effects on these cells. Furthermore, knockdown of TGF-β ligands attenuated fibroblast-mediated collagen gel contraction, and diminished lung cancer cell invasion in three-dimensional co-culture. We also observed differential effects by targeting different TGF-β isoforms in lung cancer cells and fibroblasts. Conclusions: Our findings support the notion that RNAi-mediated targeting of TGF-β ligands may be beneficial for lung cancer treatment via its action on both cancer and stromal cells. This study further demonstrates the usefulness of this three-dimensional co-culture model to examine the effect of therapeutic agents on tumor–stromal interaction. Keywords: RNA interference, MicroRNA, Lentivirus vector, TGF-β, Three-dimensional co-culture, Gel contraction assay * Correspondence: [email protected] Department of Respiratory Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Division for Health Service Promotion, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan Full list of author information is available at the end of the article © 2014 Horie et al.; licensee BioMed Central Ltd. This is an Open Access article distributed under the terms of the Creative Commons Attribution License (http://creativecommons.org/licenses/by/4.0), which permits unrestricted use, distribution, and reproduction in any medium, provided the original work is properly credited. The Creative Commons Public Domain Dedication waiver (http://creativecommons.org/publicdomain/zero/1.0/) applies to the data made available in this article, unless otherwise stated. Horie et al. BMC Cancer 2014, 14:580 Page 2 of 11 http://www.biomedcentral.com/1471-2407/14/580 Background ligands have successfully ameliorated outcomes in dis- Lung cancer causes the deaths of more than one million ease models [16], and raised hope that this approach people worldwide every year [1]. Despite recent progress may be useful in a clinical setting. in molecular-targeted therapeutics, such as inhibitors of However, the three isoforms of TGF-β ligands—TGF- epidermal growth factor receptor (EGFR) tyrosine kinase β1, TGF-β2, and TGF-β3—show different expression pro- and anaplastic lymphoma kinase (ALK), failure to achieve files in various tissues and cell types. To develop effective long-lasting therapeutic responses has emphasized the therapeutic strategies for silencing TGF-β ligands, iden- need for novel treatment strategies [2,3]. tifying the appropriate isoform and target cell type may Most forms of cancer are associated with a stromal be critical. To our knowledge, the differential effects of response and extracellular matrix (ECM) deposition, eliminating specific TGF-β isoforms in a given tissue referred to as desmoplasia, which is critically regulated type remain unstudied. by cancer-associated fibroblasts (CAFs) [4]. Cancer tissue In the present study, we explored the therapeutic remodeling allows tumor cells to grow and disseminate, effect of TGF-β signaling blockade in lung cancer. and contributes to increased interstitial fluid pressure, We previously developed a three-dimensional (3D) which can be an obstacle to drug delivery [5]. co-culture model for evaluation of tumor–stromal inter- Among the soluble factors involved in the tumor–stro- actions [17]. Using this model, we tested the differential mal interaction, transforming growth factor (TGF)-β plays effects of silencing TGF-β ligands in A549 lung cancer a pivotal role. In premalignant stages, TGF-β acts as a cells and HFL-1 lung fibroblasts. Among the three iso- tumor suppressor by inhibiting proliferation and apoptotic forms of TGF-β ligands, TGF-β1 and TGF-β2 (but not induction in epithelial cells. In later stages, epithelial cells TGF-β3) are dominantly expressed in these cells [18-20]. become refractory to the growth inhibitory effect of TGF- Thus we established lentiviral vectors that transduce artifi- β and begin to secrete high levels of TGF-β, which in turn cial miRNAs against human TGF-β1and TGF-β2 as a tool exhibits tumor-promoting activity, such as angiogenesis, for testing the effects of TGF-β ligand knockdown. immune evasion, fibroblast activation, and ECM accumu- lation [6-8]. Furthermore, TGF-β increases the migratory Methods and invasive capacity of cancer cells by inducing the Cell culture epithelial–mesenchymal transition (EMT) [9,10]. Indeed, Tissue culture media and supplements were purchased TGF-β levels in both serum and tissues were elevated and from GIBCO (Life Technologies, Grand Island, NY). associated with worsening prognosis in patients with lung A549 human lung adenocarcinoma cells and HFL-1 cancer [11,12]. As such, TGF-β may be a promising target human lung fibroblasts were purchased from the American for cancer therapy. However, in contrast to cancer cells, Type Culture Collection (Rockville, MD), and were cul- the role of TGF-β signaling in the tumor stroma is poorly tured in Dulbecco’s Modified Eagle’sMedium(DMEM) understood, at least partly due to technical limitations in supplemented with 10% fetal bovine serum (FBS). In detecting TGF-β signaling activation in situ. addition, 293FT cells were obtained from Invitrogen RNA interference (RNAi) has been used widely to in- (Carlsbad, CA), and cultured in 100-mm dish coated duce the potent and specific inhibition of gene expres- with collagen type I (IWAKI, Tokyo, Japan) in DMEM sion. Several variants of small regulatory RNAs are with 10% FBS and 1 mM sodium pyruvate. involved in RNAi, including synthetic double-stranded small interfering RNAs (siRNAs), RNA polymerase III Artificial miRNA sequences (pol III)-transcribed small hairpin RNAs (shRNAs), and The BLOCK-iT™ Pol II miR RNAi Expression Vector Kit endogenous or artificial microRNAs (miRNAs) that are with EmGFP (Invitrogen, Carlsbad, CA) was used for transcribed by RNA polymerase II (pol II) as pri-miRNA, RNAi experiments. The design of the expression vector and subsequently processed into mature miRNAs [13,14]. was based on the use of endogenous murine miR-155 Vectors that enable the expression of engineered miRNA flanking sequences. Artificial miRNA sequences target- sequences from Pol II promoters have been developed ing human TGF-β ligands were designed using BLOCK- [15], in which the stem sequences of an endogenous iT™ RNAi Designer (http://rnaidesigner.Invitrogen.com/ miRNA precursor are substituted with unrelated base- rnaiexpress/). Four and three pairs of sense and anti- paired sequences that target specific genes. sense oligonucleotides were designed for targeting hu- Among the therapeutic strategies for TGF-β signaling man TGF-β1 and β2, respectively (Additional file 1: inhibition, RNAi is emerging as a promising tool [13]. Table S1). Recent advances in RNAi technology are overcoming previous obstacles, such as instability in vivo, impeded Plasmid construction and preparation of viral vectors drug delivery, and undesirable off-target effects. In ani- The designed oligonucleotides were annealed, followed mal experiments, RNAi agents directed against TGF-β by ligation into the pcDNA6.2-GW/EmGFP-miR vector Horie et al. BMC Cancer 2014, 14:580 Page 3 of 11 http://www.biomedcentral.com/1471-2407/14/580 (Invitrogen), which facilitates transfer into a suitable des- The optical density of each reaction was measured at tination vector via Gateway recombination reactions. 450 nm using a microplate reader (Bio-Rad, Hercules, CA), The EmGFP forward sequence primer (5′- GGCATG- and corrected against absorption at 570 nm. The GACGAGCTGTACAA −3′) was used for sequencing of data were analyzed using the Microplate Manager III the miRNA insert fragments, which was performed Macintosh data analysis software (Bio-Rad). using an ABI PRISM® 310 Genetic Analyzer. As the con- trol, pcDNA6.2-GW/EmGFP-miR negative control plas- Cell proliferation assay mid (Invitrogen) was used. The sequence containing the A549 cells were seeded at a density of 1 × 10 /well on miRNA coding region was transferred to the lentivirus 12-well dishes and HFL-1 cells were seeded at 4 × 10 / vector via the Gateway cloning system (Invitrogen). well on 6-well dishes. Both cell types were cultured in Briefly, the miRNA coding region was subcloned into DMEM containing 10% FBS. Cells were counted on days the entry plasmid pDONR221 (Invitrogen) using Gateway® 1, 3, and 5 after seeding using a hemocytometer. BP Clonase™ II Enzyme Mix (Invitrogen). The sequences in the entry plasmids were then transferred to the lenti- Collagen gel contraction assay and 3D co-culture viral expression vector, pCSII-EF-RfA, using Gateway® LR Three-dimensional gel cultures were carried out accord- Clonase™ II Enzyme Mix (Invitrogen). ing to the previously published protocol [17]. Briefly, collagen gels were prepared by mixing 0.5 mL of fibro- Lentivirus infection blast cell suspension (~2.5 × 10 cells) in FBS, 2.3 mL of The recombinant lentivirus was produced by transfec- type I collagen (Cell matrix type IA; Nitta Gelatin, tion of 293FT cells with the lentiviral expression vec- Tokyo, Japan), 670 μLof5×DMEM, and330 μLofre- tors, pCMV-VSV-G-RSV-Rev, and pCAG-HIVgp, using constitution buffer, following the manufacturer’srec- Lipofectamine 2000 reagent (Invitrogen). After 72 h, the ommendations. The mixture (3 mL) was cast into each medium was collected, and 1 × 10 of A549 or HFL-1 well of the six-well culture plates. The solution was cells were infected with 500 μL of medium containing then allowed to polymerize at 37°C for 30 min. After lentiviruses. For double knockdown of TGF-β1and overnight incubation, each gel was detached and cultured TGF-β2, 250 μL of each lentivirus-containing medium in growth medium, and the surface area of the gels was were used. Infection efficiency was assessed by measur- quantified via densitometry (Densitograph, ATTO, Tokyo, ing the percentage of EmGFP-positive cells via flow cy- Japan) for 5 consecutive days, and the final size relative to tometry (EPICS XL System II; Beckman Coulter, Brea, initial size was determined. For 3D co-culture, A549 cells CA), and knockdown efficiency of target gene was ana- (2 × 10 ) were seeded on the surface of each gel prior to lyzed using an enzyme-linked immunosorbent assay overnight incubation. After 5 days of floating culture, the (ELISA). gel was fixed in formalin solution and embedded in paraf- fin, and vertical sections were stained with hematoxylin RT-PCR and eosin. Total RNA was extracted using the RNeasy Mini Kit (Qiagen, Tokyo, Japan). The cDNA was synthesized Statistics using SuperScript III Reverse Transcriptase (Invitrogen), Results were confirmed by performing experiments in following the manufacturer’s protocol. Quantitative triplicate. Analyses were performed using JMP version reverse transcription (RT)-PCR was performed using 9 (SAS Institute Inc., Tokyo, Japan). For statistical sig- Mx-3000P (Stratagene, La Jolla, CA) and QuantiTect nificance, differences between two experimental groups SYBR Green PCR (Qiagen). Relative mRNA expression were examined using Student’s t-test, and Dunnett’stest was calculated using the ΔΔC method, and expression was used for multiple comparisons with control group. was normalized to that of the glyceraldehyde 3-phosphate P< 0.05 was considered to indicate significance. dehydrogenase (GAPDH) gene. The specific primers are shown in Additional file 2: Table S2. GSEA (gene set enrichment analysis) Navab et al. reported gene expression profiles for 15 pairs ELISA for TGF-β1 and TGF-β2 of lung CAFs and NFs, and identified genes enriched in A549 and HFL-1 cells were serum-starved for 24 h, and lung CAFs [21]. GSEA was performed using these micro- each supernatant was collected. The concentrations of array data sets (GSE22862) deposited in the public data- TGF-β1and TGF-β2 were measured using the Quanti- base. To obtain a gene set regulated by TGF-β,we used kine ELISA for human TGF-β1/TGF-β2(R&DSystems, publicly available microarray datasets, derived from two Minneapolis, MN), according to the manufacturer’sin- lung fibroblast cell lines stimulated by TGF-β:HFL-1 structions. Each supernatant was activated by 1 N HCl, (GSE27597) and IMR-90 (GSE17518) [22,23]. We ex- followed by neutralization with 1.2 N NaOH/0.5 M HEPES. tracted the top 800 TGF-β-induced genes from each Horie et al. BMC Cancer 2014, 14:580 Page 4 of 11 http://www.biomedcentral.com/1471-2407/14/580 dataset, as identified through the Significance Analysis enriched in CAFs, suggesting that TGF-β signaling is ac- of Microarrays (SAM) method. Combining these two tivated in lung CAFs (Figure 1A). We further extracted gene lists, we isolated 196 commonly induced genes in 88 ‘leading edge genes’ out of the TGF-β-regulated two lung fibroblast cell lines, which were defined as genes. A heatmap of these leading edge genes clearly ‘TGF-β-regulated genes’ (Additional file 3: Table S3). illustrated differential expression between CAFs and NFs (Figure 1B). As expected, ECM-related genes were Results enriched among the leading edge genes, and a heatmap of TGF-β signaling is activated in lung CAFs 16 selected ECM related genes apparently showed that CAFs are a major constituent of the tumor stroma, and TGF-β-regulated ECM-related enzymes and substrates, we have previously shown that lung CAFs are more potent including PLOD1, LOX,COL1A1, VCAN,SPARC,FN1, in enhancing cancer cell invasion and collagen gel con- ELN,and THBS1, aremoreenrichedinCAFsthanNFs traction than normal lung fibroblasts (NFs) [17]. Although (Figure 1C). theroleofTGF-β in cancer cells and lung fibroblasts has been investigated extensively, TGF-β function in CAFs Lentivirus-mediated transduction of artificial miRNAs remains largely unknown due to technical hurdles in against human TGF-β1 and TGF-β2 isolating fibroblasts from lung cancer tissues. Based on the observation that endogenous TGF-β signal- To examine TGF-β signaling activation status in lung ing is activated in lung CAFs, we examined whether TGF- CAFs, we used gene set enrichment analysis (GSEA) to β signaling activation in fibroblasts modulates the behavior determine whether the expression of the identified TGF- of adjacent cancer cells. We also aimed to elucidate the β-regulated genes was enhanced in lung CAFs compared cell-autonomous action of TGF-β in lung cancer cells. To to NFs. This was performed using microarray data sets this end, we generated lentiviral vectors that transduced of CAFs and NFs reported by Navab et al. [21]. These artificial miRNAs against TGF-β ligands, and tested their analyses demonstrated that the TGF-β-regulated genes effects on lung cancer cells and HFL-1 lung fibroblasts. identified through our analysis are in fact highly The expression levels of TGF-β isoforms are variable Figure 1 Gene set enrichment analysis (GSEA). A: GSEA was used to examine the enrichment of identified TGF-β-regulated genes in CAFs. ‘TGF-β-regulated genes’ include 196 genes induced by TGF-β in both IMR-90 and HFL-1 lung fibroblast cell lines. CAF and NF gene expression profiles reported by Navab et al. [21] were used. Enrichment of TGF-β-regulated genes is shown schematically with those that best correlated with the CAF phenotype on the left (‘CAF-high’) and the genes that best correlated with the NF phenotype on the right (‘NF-high’). B: A heat map representing the relative expression change of ‘ 88 leading edge genes’ which were obtained by GSEA analysis in CAFs and NFs. C: A heat map representing the relative expression change of selected ‘16 ECM related genes’. Horie et al. BMC Cancer 2014, 14:580 Page 5 of 11 http://www.biomedcentral.com/1471-2407/14/580 among lung cancer cell lines. In order to survey these dif- against TGF-β1 were ineffective for TGF-β2 (Figure 2B, ferences, we used Cancer Cell Line Encyclopedia (CCLE) right). These results show that miRNAs against TGF-β1 data and found that expression of TGF-β isoforms are or TGF-β2 exert their effects in a selective manner for relatively high in A549 cells among 111 non-small cell each ligand. Out of the two combinations tested for lung cancer cell lines (Additional file 4: Figure S1). There- double knockdown, miRNA #2 against TGF-β1and #2 fore, we used A549 lung cancer cells in the following against TGF-β2 showed efficient silencing in both A549 experiments. and HFL-1 cells (Figure 2). Therefore, we selected miRNA Four miRNA sequences were designed to target hu- sequences #2 against TGF-β1 and #2 against TGF-β2, for man TGF-β1, as well as three sequences against TGF-β2 single or double knockdown in the following experiments. (Additional file 1: Table S1). Next, we determined the ef- ficiency of lentiviral infection by measuring the percent- Cell proliferation is suppressed by knockdown of TGF-β1 age of EmGFP-positive cells using flow cytometry. More and/or TGF-β2 than 95% of A549 cells were positive for EmGFP, sug- Next, we investigated whether TGF-β1 and/or TGF-β2 gesting a high transduction efficiency for this miRNA knockdown affected the proliferation of A549 and HFL-1 sequence (Additional file 5: Figure S2A, left); we ob- cells. In both cell types, the transduction of artificial served similar efficiencies for all miRNA sequences used miRNAs against TGF-β1or TGF-β2 suppressed cell in this study (Additional file 5: Figure S2A, right). proliferation (Figure 3), and this anti-proliferative effect Meanwhile, HFL-1 cells showed more modest (but still was enhanced in cells subject to double knockdown, com- sufficient) efficiencies for lentiviral infection (Additional pared to single knockdown of either TGF-β1orTGF-β2. file 5: Figure S2B, left). The percentage of EmGFP- TGF-β is a strong inhibitor of proliferation in most positive cells ranged from 65–85% among the miRNA epithelial cells, whereas it promotes proliferation in mes- sequences (Additional file 5: Figure S2B, right). enchymal cells and enhances cancer cell survival [6-8]. For double knockdown of TGF-β1 and TGF-β2, two Our lentivirus-mediated miRNA delivery system main- combinations of lentiviruses encoding miRNAs against tains stable knockdown of TGF-β1 and/or TGF-β2. This TGF-β1and TGF-β2 were co-infected: #2 miRNA against may alter cell signaling in the steady state and modulate TGF-β1and #2 miRNA against TGF-β2 (TGF-β1KD #2+ the cell machinery that regulates cell survival or prolifer- TGF-β2KD #2), or #4 miRNA against TGF-β1and #3 ation, thereby resulting in suppressed cell proliferation. miRNA against TGF-β2(TGF-β1KD #4+ TGF-β2KD #3). Co-infection with two different lentiviruses showed similar transduction efficiencies compared to single infections, as Altered EMT-related gene expression via TGF-β1 and/or determined via EmGFP fluorescence (Additional file 5: TGF-β2 knockdown Figure S2A, right and Additional file 5: Figure S2B, right). EMT is crucial for cancer cells to acquire invasive pheno- types, which are characterized by downregulation of E- Potent and selective knockdown of TGF-β1 and TGF-β2 cadherin and upregulation of vimentin. A549 cells stay in Next, we evaluated the efficiency of TGF-β knockdown an intermediary state of EMT, whereas exogenous TGF-β through measurement of protein expression via ELISA. further promotes acquisition of mesenchymal phenotypes To control for unintended effects of experimental ma- [20]. We examined whether knockdown of TGF-β ligands nipulation, we examined the expression of TGF-β1 and modulated the expression of EMT markers. TGF-β2 in uninfected A549 and HFL-1 cells compared Silencing of TGF-β2 led to E-cadherin upregulation, to cells infected with negative control (NTC) miRNAs suggesting the restoration of epithelial phenotypes. In ac- (Figure 2). No significant difference in TGF-β1orTGF-β2 cordance, vimentin expression was suppressed by knock- expression was observed. down of TGF-β1and/or TGF-β2, though it failed to reach In A549 cells, three of four miRNAs against TGF-β1 statistical significance (Figure 4A). These results support (#1, #2, and #4) were able to silence TGF-β1 expression, the notion that endogenous TGF-β signaling participates whereas all three miRNAs against TGF-β2 were ineffective in the maintenance of a mesenchymal phenotype in A549 for TGF-β1 (Figure 2A, left). Two out of three miRNAs cells in the steady state. against TGF-β2 (#2 and #3) silenced TGF-β2 expression, EMT is accompanied by the enhanced expression whereas all four miRNAs against TGF-β1 were ineffective of fibrogenic growth factors, such as platelet-derived for TGF-β2 (Figure 2A, right). In HFL-1 cells, three of growth factor (PDGF) and connective tissue growth fac- four miRNAs against TGF-β1 (#1, #2 and #3) were able to tor (CTGF) [20]. PDGF is a dimeric protein composed silence TGF-β1 expression, whereas all three miRNAs of A and B subunits, and it has been reported that the against TGF-β2 were ineffective for TGF-β1 (Figure 2B, transcription of PDGFB is regulated by TGF-β.Consist- left). Two of three miRNAs against TGF-β2 (#2 and #3) si- ent with the previous experiment [20], TGF-β2 silencing lenced TGF-β2 expression, whereas all four miRNAs led to CTGF downregulation, whereas knockdown of Horie et al. BMC Cancer 2014, 14:580 Page 6 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 2 Knockdown of TGF-β ligands. A: TGF-β1 and TGF-β2 concentrations measured by ELISA in the supernatant of A549 cells transduced with each miRNA. Left: TGF-β1. Right: TGF-β2. Data shown are the means ± SEM of triplicate analyses. KD: knockdown. NTC: negative control. The concentration of TGF-β1 or TGF-β2 in the supernatant of cells with TGF-β1 and/or TGF-β2 knockdown was compared to that of cells transduced with NTC miRNA. Statistical significance was determined by Dunnett’s test. P< 0.05. B: TGF-β1 and TGF-β2 concentrations measured by ELISA in the supernatant of HFL-1 cells. TGF-β1 and/or TGF-β2 attenuated PDGFB expression TGF-β1 and/or TGF-β2 knockdown attenuates collagen (Figure 4B). gel contraction in HFL-1 cells Upon TGF-β stimulation, fibroblasts convert to an acti- Cancer tissue contraction facilitates tumor progression vated phenotype to enhance ECM production. Thus, we and contributes to increased interstitial fluid pressure, examined whether knockdown of TGF-β1and/orTGF-β2 which hampers drug delivery [5]. The collagen gel con- modulated the expression of α1 (I) collagen (COL1A1), a traction assay is used widely to recreate tissue contraction major component of ECM. In HFL-1 cells, TGF-β1 knock- in an experimental setting, and it has been shown that down decreased the expression of COL1A1, whereas TGF-β stimulates fibroblast-mediated collagen gel con- TGF-β2 silencing had no effect (Figure 4C). traction [25]. We used this assay to investigate whether These results suggest the differential regulation of knockdown of TGF-β1 and/or TGF-β2 modulated tissue target genes by TGF-β1orTGF-β2 in cancer cells and contraction through effects on fibroblasts. fibroblasts. During lung branching morphogenesis, Collagen gels were embedded with HFL-1 cells after TGF-β1 expression is prominent throughout the mes- TGF-β1 and/or TGF-β2 knockdown, and gel size was enchyme, whereas TGF-β2 is localized to mainly the measured daily. On the first day, the control gel size was epithelium of the developing distal airways [24]. Thus, reduced to ~50% of the initial value, followed by gradual TGF-β2 may be critical for determining epithelial or shrinkage to less than 20% on the fifth day (Figure 5). cancer cell behavior in a cell-autonomous fashion, Compared to the control, knockdown of TGF-β1and/or whereas endogenous TGF-β1 may play a greater role in TGF-β2 in HFL-1 cells attenuated gel contraction (Figure 5 fibroblasts. and Additional file 6: Figure S3). These results suggested Horie et al. BMC Cancer 2014, 14:580 Page 7 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 3 Cell proliferation assay. Cell proliferation curve in A549 or HFL-1 cells transduced with NTC miRNA (solid line) compared to cells transduced with miRNA against TGF-β1 (dashed line: TGF-β1KD),TGF-β2(dotted line: TGF-β2KD),orTGF-β1 and TGF-β2 (dashed-dotted line: TGF-β1+ β2KD).Cell counts were carried out on days 1, 3, and 5 after seeding. Left: A549. Right: HFL-1. Data shown are the means ± SEM of triplicate analyses. Numbers of cells with TGF-β1and/orTGF-β2 knockdown on day 5 was compared to that in the cells transduced with NTC miRNA. Statistical significance was determined by Student’s t-test. P< 0.05. that the inhibition of endogenous TGF-β signaling in Discussion fibroblasts ameliorates tissue contraction. TGF-β plays several crucial roles in cancer progression, affecting both tumor and stromal cells, including fi- broblasts [4]. However, very little is known regarding Three-dimensional co-culture of A549 and HFL-1 cells the effects of TGF-β ligand silencing in the context To examine the interaction between lung cancer cells of tumor–stromal or epithelial–mesenchymal interac- and fibroblasts, we previously established a 3D co-culture tions [26]. Numerous reports have shown the effects model [17]. HFL-1 cells transduced with control miRNAs of exogenous TGF-β stimulation in various cell types, or those for TGF-β1and TGF-β2 silencing (double knock- whereas the effects of endogenous or cell-autonomous down) were embedded into the collagen gels, and then TGF-β signaling are poorly understood. To our know- A549 cells were seeded onto the surface of these gels. The ledge, this study is the first to generate lentiviral vectors co-cultured collagen gels were subjected to floating cul- encoding artificial miRNAs targeting human TGF-β1and ture for an additional 5 days, followed by hematoxylin and TGF-β2, and to explore their effects in a co-culture eosin staining (Figure 6). model. Double knockdown of TGF-β1 and TGF-β2 in HFL-1 Lentiviral vectors showed efficient transduction in A549 cells did not show clear effects on A549 cell invasion, lung cancer cells, as well as HFL-1 lung fibroblasts. suggesting a minor role for TGF-β produced in HFL-1 Knockdown efficiency to less than 30% of the control was cells in this co-culture model (lower panels). In our obtained for both TGF-β1and TGF-β2 in a selective previous work, we did not examine whether HFL-1 manner. Knockdown of TGF-β ligands suppressed cell cells enhance lung cancer cell invasion [17], and this proliferation in both A549 and HFL-1 cells. Furthermore, study suggests that endogenous TGF-β expression in expression of EMT markers and fibrogenic growth factors HFL-1 cells may not have a significant role in invasion was modulated in A549 cells, whereas collagen I was promotion. downregulated in HFL-1 cells. With regard to cellular In contrast, A549 cell invasion was observed when function, silencing of TGF-β ligands attenuated HFL-1- control A549 cells were cultured with control HFL-1 mediated collagen gel contraction, and inhibited A549 cell cells (upper left panel). Silencing of either TGF-β1or invasion in the 3D co-culture model. All of these findings TGF-β2 in A549 cells failed to inhibit invasion (upper support the tumor-promoting role of TGF-β,and that middle panels), whereas double knockdown of TGF-β1 the reported beneficial effects of TGF-β inhibition in and TGF-β2 led to complete disappearance of invading cancer therapeutics may derive from interfering with cells (upper right panel). tumor–stromal communications. Horie et al. BMC Cancer 2014, 14:580 Page 8 of 11 http://www.biomedcentral.com/1471-2407/14/580 Figure 4 Quantitative RT-PCR. A: Quantitative RT-PCR for E-cadherin (left) and vimentin (right) in A549 cells. B: Quantitative RT-PCR for CTGF (left) and PDGFB (right) in A549 cells. C: Quantitative RT-PCR for COL1A1 in HFL-1 cells. Data shown are the means ± SEM. The relative expression of each gene in cells with TGF-β1 and/or TGF-β2 knockdown was compared to that in the cells transduced with NTC miRNA. Statistical significance was determined by Student’s t-test. P< 0.05. In our experiments, it appeared that both TGF-β1and and invasion in a 3D co-culture. In HFL-1 cells, TGF-β1 TGF-β2wereabundantlyproducedinA549 cells,whereas knockdown was more effective than TGF-β2knockdown the concentration of TGF-β1 was higher than that of in suppressing COL1A1 expression. TGF-β2 in the supernatant of HFL-1 cells. Compared to Little is known regarding the expression profiles of single knockdown, double knockdown of TGF-β1and TGF-β isoforms in various lung cancer cell types. As TGF-β2 showed stronger effects in A549 cell proliferation shown here, knockdown of each TGF-β ligand Horie et al. BMC Cancer 2014, 14:580 Page 9 of 11 http://www.biomedcentral.com/1471-2407/14/580 knockdown of TGF-β1 and TGF-β2 in HFL-1 cells did not show clear effects on A549 cell invasion, and en- NTC miRNA dogenous TGF-β expression in HFL-1 cells seemed to TGF- 1 KD have little effect on lung cancer cell invasion. The pre- TGF- 2 KD cise mechanism underlying CAF-enhanced lung cancer TGF- 1+ 2KD cell invasion remains to be elucidated, and further stud- ies are necessary to clarify the mechanisms underlying cell invasion in our experimental model. There have been several attempts to exploit TGF-β signaling inhibition as a therapeutic approach for malig- nant tumors, including the use of TGF-β receptor kinase inhibitors, TGF-β neutralizing antibodies, TGF-β anti- sense oligonucleotides (AONs), and siRNAs [27]. TGF-β type I receptor kinase inhibitor has been tested for non- small cell lung cancer (NSCLC) patients in a phase II study, but failed to yield clinical benefits [28]. Several animal models of cancer have demonstrated the thera- * peutic effect of TGF-β neutralizing antibodies [29]. Recently, AONs against TGF-β ligands have shown promising clinical results. Trabedersen (AP 12009) is an AON against human TGF-β2. Intra-tumoral administra- tion of trabedersen in patients with high-grade gliomas 01 2 3 4 5 led to better tumor control and prolonged survival with Day fewer adverse events, which prompted a larger phase III Figure 5 Collagen gel contraction assay. Time-course of gel trial [30]. Intravenous application of trabedersen in pa- contraction in the presence of HFL-1 transduced with NTC miRNA tients with other cancer types is also under evaluation. AP (solid line), or miRNAs against TGF-β1 (dashed line: TGF-β1KD), TGF-β2 11014, another AON targeting human TGF-β1, is cur- (dotted line: TGF-β2KD), orTGF-β1 and TGF-β2 (dashed-dotted line: rently in preclinical development for NSCLC treatment. TGF-β1+ β2 KD). The area of each gel was assessed daily for 5 days Furthermore, a phase II trial for belagenpumatucel-L, a and the relative value compared to the initial size was determined. Data shown are the means ± SD of triplicate analyses. Statistical vaccine produced from NSCLC cells transfected with significance was determined by Student’st-test. *P< 0.05. TGF-β2 AON, has shown beneficial effects on survival without any significant adverse effects; phase III studies in modulated phenotype in a cell-type-dependent manner. lung cancer patients are ongoing [31]. RNAi targeting These effects may be much more complicated and variable TGF-β ligands is also emerging as a promising tool [13]. depending on the multicellular context; nevertheless, our In animal experiments, RNAi agents against TGF-β1 results demonstrate the important role for TGF-β signal- demonstrated therapeutically beneficial effects, support- ing in the tumor microenvironment. ing progression toward future clinical applications [16]. We have reported previously that lung CAFs enhance This body of work demonstrates the intensifying interest cancer cell invasion [17]. In the present study, double in TGF-β ligand silencing as a therapeutic approach for Figure 6 3D co-culture model. Hematoxylin and eosin staining of 3D cultured gels composed of A549 and HFL-1 cells transduced with the indicated miRNAs. Upper panels: HFL-1 cells transduced with NTC miRNA. Lower panels: HFL-1 cells transduced with miRNAs against TGF-β1 and TGF-β2 (TGF-β1+ β2 KD). Invading cells are indicated with arrows. Scale bar: 100 μm. % of initial size of gel Horie et al. BMC Cancer 2014, 14:580 Page 10 of 11 http://www.biomedcentral.com/1471-2407/14/580 lung cancer. To validate therapeutic strategies against Authors’ contributions MH carried out the experiments and drafted the manuscript. AS, TK, and TN TGF-β ligands, it may be critical to target the appropriate designed the study and participated in manuscript preparation. SN and HIS TGF-β isoform in a given cell type. The present study pro- performed statistical analyses. MO participated in the design of the study. vides a useful experimental model to investigate the effect YM participated in preparation of tissue sections. All authors read and approved the final manuscript. of therapeutic agents targeting TGF-β ligands. Our results suggest that targeting both TGF-β1and TGF-β2inlung Acknowledgements cancer cells is more effective than single knockdown. Fur- This work was supported by KAKENHI (Grants-in-Aid for Scientific Research) thermore, TGF-β2 knockdown may play a more specific from the Ministry of Education, Culture, Sports, Science, and Technology, and a grant to the Respiratory Failure Research Group from the Ministry of role in lung cancer cells than in stromal cells, such as fi- Health, Labour and Welfare, Japan. We thank Makiko Sakamoto for the broblasts. Future studies are warranted to further elucidate technical assistance. the therapeutical benefits of strategies against the different Author details TGF-β ligands. Department of Respiratory Medicine, Graduate School of Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, Japan. Division for Health Service Promotion, The University of Tokyo, 7-3-1 Hongo, Conclusion Bunkyo-ku, Tokyo 113-0033, Japan. Department of Biochemistry, Nihon Because TGF-β exerts it pleiotropic effects in a variety University School of Dentistry, 1-8-13 Kanda-Surugadai, Chiyoda-ku, Tokyo of cells in the tumor microenvironment, it is useful to 101-8310, Japan. Department of Biochemistry, Ohu University School of Pharmaceutical Sciences, Misumido 31-1, Tomitamachi, Koriyama, Fukushima evaluate the action of anti-TGF-β therapeutic agents in 963-8611, Japan. Department of Molecular Pathology, Graduate School of multicellular culture conditions. Our 3D co-culture Medicine, The University of Tokyo, 7-3-1 Hongo, Bunkyo-ku, Tokyo 113-0033, model, demonstrated here, represents a useful tool for Japan. The Fourth Department of Internal Medicine, Teikyo University School of Medicine University Hospital, Mizonokuchi, 3-8-3 Mizonokuchi, evaluating differential effects on cancer cells and fibro- Takatsu-ku, Kawasaki, Kanagawa 213-8507, Japan. blasts. In summary, we established a lentivirus-mediated knockdown system for TGF-β ligands, which revealed Received: 8 February 2014 Accepted: 4 August 2014 Published: 9 August 2014 their multifaceted effects on cell proliferation, EMT, inva- sion, and ECM remodeling. References 1. Jemal A, Bray F, Center MM, Ferlay J, Ward E, Forman D: Global cancer statistics. CA Cancer J Clin 2011, 61(2):69–90. Additional files 2. Maemondo M, Inoue A, Kobayashi K, Sugawara S, Oizumi S, Isobe H, Gemma A, Harada M, Yoshizawa H, Kinoshita I, Fujita Y, Okinaga S, Hirano H, Additional file 1: Table S1. Sequences of artificial miRNAs against Yoshimori K, Harada T, Ogura T, Ando M, Miyazawa H, Tanaka T, Saijo Y, TGF-β ligands. Hagiwara K, Morita S, Nukiwa T: Gefitinib or chemotherapy for non-small- cell lung cancer with mutated EGFR. N Engl J Med 2010, Additional file 2: Table S2. Primers for RT-PCR. 362(25):2380–2388. Additional file 3: Table S3. The 196 ‘TGF-β-regulated genes’. 3. Choi YL, Soda M, Yamashita Y, Ueno T, Takashima J, Nakajima T, Yatabe Y, Additional file 4: Figure S1. Expression levels of TGF-β isoforms in Takeuchi K, Hamada T, Haruta H, Ishikawa Y, Kimura H, Mitsudomi T, Tanio non-small cell lung cancer cell lines. The transcription levels of TGF-β1 Y, Mano H: EML4-ALK mutations in lung cancer that confer resistance to and TGF-β2 in non-small cell lung cancer cell lines were retrieved from ALK inhibitors. N Engl J Med 2010, 363(18):1734–1739. Cancer Cell Line Encyclopedia (CCLE) database and shown in a scatter 4. Micke P, Ostman A: Tumour-stroma interaction: cancer-associated plot. A549 cells showed relatively higher levels of TGF-β1 and TGF-β2. fibroblasts as novel targets in anti-cancer therapy? Lung Cancer 2004, 45(Suppl 2):S163–S175. Additional file 5: Figure S2. Transduction efficiency of lentiviral 5. Heldin CH, Rubin K, Pietras K, Ostman A: High interstitial fluid pressure - vectors. A: Transduction efficiency of miRNAs in A549 cells. Left: miRNA an obstacle in cancer therapy. Nat Rev Cancer 2004, 4(10):806–813. transduction was tracked by detecting EmGFP-positive cells using the 6. Massagué J: TGFbeta in Cancer. Cell 2008, 134(2):215–230. FL-1 channel of a flow cytometer. A representative result of #2 miRNA 7. Bierie B, Moses HL: Tumour microenvironment: TGFbeta: the molecular transduction against TGF-β1 is shown. The grey and black peaks are from Jekyll and Hyde of cancer. Nat Rev Cancer 2006, 6(7):506–520. uninfected and lentivirus-transduced cells, respectively. Right: transduction 8. Ikushima H, Miyazono K: TGFbeta signalling: a complex web in cancer efficiency of each miRNA. KD: knockdown. NTC: negative control. B: progression. Nat Rev Cancer 2010, 10(6):415–424. Transduction efficiency of miRNA in HFL-1 cells. 9. Saito A, Suzuki HI, Horie M, Ohshima M, Morishita Y, Abiko Y, Nagase T: An Additional file 6: Figure S3. Collagen gel contraction assay. integrated expression profiling reveals target genes of TGF-beta and Photographs of the gels on day 5 in the experiments shown in Figure 5. TNF-alpha possibly mediated by microRNAs in lung cancer cells. PLoS Identically sized white circles in each well are shown to demonstrate the One 2013, 8(2):e56587. differences in gel size. 10. Kalluri R, Weinberg RA: The basics of epithelial-mesenchymal transition. J Clin Invest 2009, 119(6):1420–1428. 11. Kong F, Jirtle RL, Huang DH, Clough RW, Anscher MS: Plasma transforming Abbreviations growth factor-beta1 level before radiotherapy correlates with long term TGF-β: Transforming growth factor-β; EGFR: Epidermal growth factor outcome of patients with lung carcinoma. Cancer 1999, 86(9):1712–1719. receptor; ALK: Anaplastic lymphoma kinase; CAFs: Cancer-associated 12. Hasegawa Y, Takanashi S, Kanehira Y, Tsushima T, Imai T, Okumura K: fibroblasts; EMT: Epithelial-mesenchymal transition; ECM: Extracellular matrix; Transforming growth factor-beta1 level correlates with angiogenesis, GSEA: Gene set enrichment analysis; PDGF: Platelet-derived growth factor; tumor progression, and prognosis in patients with nonsmall cell lung CTGF: Connective tissue growth factor; RNAi: RNA interference; carcinoma. Cancer 2001, 91(5):964–971. AON: Antisense oligonucleotides. 13. Davidson BL, McCray PB Jr: Current prospects for RNA interference-based therapies. Nat Rev Genet 2011, 12(5):329–340. Competing interests 14. Cullen BR: Transcription and processing of human microRNA precursors. The authors declare that they have no competing interests. Mol Cell 2004, 16(6):861–865. Horie et al. BMC Cancer 2014, 14:580 Page 11 of 11 http://www.biomedcentral.com/1471-2407/14/580 15. Zeng Y, Wagner EJ, Cullen BR: Both natural and designed micro RNAs can TGF-beta2 inhibitor trabedersen: results of a randomized and controlled inhibit the expression of cognate mRNAs when expressed in human phase IIb study. Neuro-Oncology 2011, 13(1):132–142. cells. Mol Cell 2002, 9(6):1327–1333. 31. Nemunaitis J, Dillman RO, Schwarzenberger PO, Senzer N, Cunningham C, 16. Hamasaki T, Suzuki H, Shirohzu H, Matsumoto T, D'Alessandro-Gabazza CN, Cutler J, Tong A, Kumar P, Pappen B, Hamilton C, DeVol E, Maples PB, Liu L, Gil-Bernabe P, Boveda-Ruiz D, Naito M, Kobayashi T, Toda M, Mizutani T, Chamberlin T, Shawler DL, Fakhrai H: Phase II study of belagenpumatucel-L, a Taguchi O, Morser J, Eguchi Y, Kuroda M, Ochiya T, Hayashi H, Gabazza EC, transforming growth factor beta-2 antisense gene-modified allogeneic Ohgi T: Efficacy of a novel class of RNA interference therapeutic agents. tumor cell vaccine in non-small-cell lung cancer. JClin Oncol 2006, PLoS One 2012, 7(8):e42655. 24(29):4721–4730. 17. Horie M, Saito A, Mikami Y, Ohshima M, Morishita Y, Nakajima J, Kohyama T, Nagase T: Characterization of human lung cancer-associated fibroblasts doi:10.1186/1471-2407-14-580 in three-dimensional in vitro co-culture model. Biochem Biophys Res Cite this article as: Horie et al.: Differential knockdown of TGF-β ligands Commun 2012, 423(1):158–163. in a three-dimensional co-culture tumor- stromal interaction model of lung cancer. BMC Cancer 2014 14:580. 18. Wen FQ, Kohyama T, Skold CM, Zhu YK, Liu X, Romberger DJ, Stoner J, Rennard SI: Glucocorticoids modulate TGF-beta production by human fetal lung fibroblasts. Inflammation 2003, 27(1):9–19. 19. Jakowlew SB, Mathias A, Chung P, Moody TW: Expression of transforming growth factor beta ligand and receptor messenger RNAs in lung cancer cell lines. Cell Growth Differ 1995, 6(4):465–476. 20. Saito RA, Watabe T, Horiguchi K, Kohyama T, Saitoh M, Nagase T, Miyazono K: Thyroid transcription factor-1 inhibits transforming growth factor- beta-mediated epithelial-to-mesenchymal transition in lung adenocarcinoma cells. Cancer Res 2009, 69(7):2783–2791. 21. Navab R, Strumpf D, Bandarchi B, Zhu CQ, Pintilie M, Ramnarine VR, Ibrahimov E, Radulovich N, Leung L, Barczyk M, Panchal D, To C, Yun JJ, Der S, Shepherd FA, Jurisica I, Tsao MS: Prognostic gene-expression signature of carcinoma-associated fibroblasts in non-small cell lung cancer. Proc Natl Acad Sci U S A 2011, 108(17):7160–7165. 22. Campbell JD, McDonough JE, Zeskind JE, Hackett TL, Pechkovsky DV, Brandsma CA, Suzuki M, Gosselink JV, Liu G, Alekseyev YO, Xiao J, Zhang X, Hayashi S, Cooper JD, Timens W, Postma DS, Knight DA, Marc LE, James HC, Avrum S: A gene expression signature of emphysema-related lung destruction and its reversal by the tripeptide GHK. Genome Med 2012, 4(8):67. 23. Hecker L, Vittal R, Jones T, Jagirdar R, Luckhardt TR, Horowitz JC, Pennathur S, Martinez FJ, Thannickal VJ: NADPH oxidase-4 mediates myofibroblast activation and fibrogenic responses to lung injury. Nat Med 2009, 15(9):1077–1081. 24. Bragg AD, Moses HL, Serra R: Signaling to the epithelium is not sufficient to mediate all of the effects of transforming growth factor beta and bone morphogenetic protein 4 on murine embryonic lung development. Mech Dev 2001, 109(1):13–26. 25. Saito RA, Micke P, Paulsson J, Augsten M, Peña C, Jönsson P, Botling J, Edlund K, Johansson L, Carlsson P, Jirström K, Miyazono K, Ostman A: Forkhead box F1 regulates tumor-promoting properties of cancer- associated fibroblasts in lung cancer. Cancer Res 2010, 70(7):2644–2654. 26. Kage H, Sugimoto K, Sano A, Kitagawa H, Nagase T, Ohishi N, Takai D: Suppression of transforming growth factor beta1 in lung alveolar epithelium-derived cells using adeno-associated virus type 2/5 vectors to carry short hairpin RNA. Exp Lung Res 2011, 37(3):175–185. 27. Hawinkels LJ, Ten Dijke P: Exploring anti-TGF-beta therapies in cancer and fibrosis. Growth Factors 2011, 29(4):140–152. 28. Scagliotti GV, Ilaria R Jr, Novello S, von Pawel J, Fischer JR, Ermisch S, de Alwis DP, Andrews J, Reck M, Crino L, Eschbach C, Manegold C: Tasisulam sodium (LY573636 sodium) as third-line treatment in patients with unresectable, metastatic non-small-cell lung cancer: a phase-II study. J Thorac Oncol 2012, 7(6):1053–1057. 29. Biswas S, Guix M, Rinehart C, Dugger TC, Chytil A, Moses HL, Freeman ML, Arteaga CL: Inhibition of TGF-beta with neutralizing antibodies prevents Submit your next manuscript to BioMed Central radiation-induced acceleration of metastatic cancer progression. J Clin and take full advantage of: Invest 2007, 117(5):1305–1313. 30. Bogdahn U, Hau P, Stockhammer G, Venkataramana NK, Mahapatra AK, Suri • Convenient online submission A, Balasubramaniam A, Nair S, Oliushine V, Parfenov V, Poverennova I, Zaaroor M, Jachimczak P, Ludwig S, Schmaus S, Heinrichs H, • Thorough peer review Schlingensiepen KH: Targeted therapy for high-grade glioma with the • No space constraints or color figure charges • Immediate publication on acceptance • Inclusion in PubMed, CAS, Scopus and Google Scholar • Research which is freely available for redistribution Submit your manuscript at www.biomedcentral.com/submit

Journal

BMC Cancer – Springer Journals

Published: Aug 9, 2014

There are no references for this article.