Tumorigenic Responses of Cancer-associated Stromal Fibroblasts after Ablative Radiotherapy: a Transcriptome-profiling Study

Cancer-associated fibroblasts (CAFs) are key elements in the progression of cancer and thereby represent important targets for cancer therapies. Increased attention has been given to ablative radiotherapy in the clinics. Therefore, in this study we have aimed at identifying the transcriptional responses occurring in primary CAFs exposed to high-dose irradiation. Established primary CAFs obtained from non-small-cell lung cancer (NSCLC) patient material were irradiated with a single dose of 18 Gy and total RNA was isolated 24 hrs after treatment. Radiation-induced transcriptional alterations were investigated by gene expression analysis using genome-wide microarrays. Obtained results were verified by qRT-PCR of relevant genes. Confirmation of gene expression outcomes was achieved by diverse functional and expression assays including DNA damage response, measurements of reactive oxygen species (ROS) by flow cytometry and senescence-associated β-galactosidase. Irradiation resulted in differential expression of 680 genes of which 557 were up-and 127 down-regulated. Of those, 153 genes were differentially expressed with a fold-change greater than 1.0 and an adjusted p-value less than 0.05 across different comparisons (non-irradiated vs. irradiated). Expression patterns revealed profound changes in biological functions and processes involved in DNA repair, apoptosis, p53 pathway, auto-phagy, senescence, ROS production and immune response. CAFs display pro-and anti-tumorigenic effects after having received a single high-dose radiation. The measured effects will have an impact on the tumor microenvironment in respect to tumor growth and metastasis.


Introduction
Hypofractionated radiotherapy is emerging as a new treatment option for solid neoplasms of different kinds, including but not limited to medically inoperable stage I non-smal-cell lung cancers (NSCLC) [1,2].Advances in radiotherapy (RT)-technology thus now permit accurate delivery of high-dose (or ablative) RT to target earlystage small tumors with acceptable toxicity to the surrounding normal tissue [3].Clinical outcomes indicate that ablative RT is associated with improved local control and overall survival compared with conventional RT regimens [2,4].Despite the encouraging clinical outcomes presented in recent years, little attention has been paid to the biological reactions associated to ablative doses of radiation to tumors.
Tumor progression is a multi-step process orchestrated by many different cell associated to the malign tissue, such as cancer-associated fibroblasts (CAFs), pericytes, lymphocytes, endothelial cells, extracelluar matrix and myeloid cells [5][6][7].CAFs are frequently found in the reactive stroma of cancers, and their presence in large number is associated with poor prognosis.When acting on cancer cells, CAFs promote tumor growth and invasion, and enhance angiogenesis by secreting factors that activate endothelial cells and pericytes.Through the secretion of cytokines, CAFs and immune cells can exert both, tumor-suppressing and tumor-promoting effects [8].Exposure of normal fibroblasts to ionizing radiation has been a common model to study cellular responses to genotoxic stress [9,10].Aiming for a better understanding of the collateral effects induced by ionizing radiation on normal tissue, some studies have tried to decode the overall changes in gene expression on normal fibroblasts by applying microarray gene technology [9][10][11][12][13].After

Irradiation of Cells
Radiation protocols were established after initial doseescalating pilot trials.Hence, adherent CAFs cultured in flasks were irradiated at RT with high energy photons produced by a Varian clinical linear-accelerator, delivered as single doses of 2, 6, 12 and 18 Gy.Standard parameters for dose delivery was depth 30 mm, beam quality 15 MV, dose-rate 6 Gy/min and field size 20 × 20 cm.Radiation-doses were confirmed to be correct within an acceptable ± 4% by Thermo-Luminescent Dosimeters (TLDs) [15].Cell survival/death after radiation was assessed by light microscopy during the initial three weeks, by automated cell index analyses (xCelligence, Roche) and by MTT cell proliferation and viability assays [16] at different time points [14].The extent of radiation-induced apoptosis and necrosis in CAFs was determined by staining for annexin V and propidium iodide respectively (data not shown).

DNA Damage Response (DDR) Foci Staining
CAFs were cultured in 2-well chamber slides (Nunc, Thermo Fisher Scientific, NY, USA), fixed with 4% PFA in PBS for 10 min at RT and permeabilized with 0.2% Triton in PBS for 8 min.Slides were then exposed to blocking buffer containing 2% HSA in PBS, for 30 min at RT.Primary antibody (Rabbit anti human 53BP1; Cat.#ab36823, Abcam, Cambridge, UK) was diluted in blocking buffer and incubated with CAFs for 45 min at RT.After washing with PBS, cells were incubated with secondary antibody (anti rabbit-Alexa546, Cat.#A11010,Molecular Probes/Invitrogen, Leiden, The Netherlands) in blocking buffer, 30 min at RT.A second wash was followed by preparation of slides in DAPI-Fluoromount-G (Cat.#0100-20,Southern Biotech, Birmingham, AL, USA).Specimens were examined in a fluorescence microscope (Zeiss Axiophot, Germany) equipped with a Nikon DS-5MC digital camera, and images were processed with Adobe ® Photoshop Software (CS5).

Senescence Associated β-Galactosidase Assays
CAFs were seeded at a density of 20,000 cells per well in uncoated 6-well plates and left for attachment for 24 hrs.Adherent cells were irradiated with a single fraction of 18 Gy.Five days post-irradiation, cultures were washed twice in PBS, and fixed for 5 min at RT with freshly prepared paraformaldehyde (2%).β-galactosidase (5bromo-4chloro-3-indolyl-B-D-galactopyranoside) staining was achieved following instructions from the manufacturer with the "Senescence Cells Histochemical Staining Kit" (Cat.noCS0030, Sigma-Aldrich, St. Louise, MO, USA).The number of β-galactosidase active and sensecent cells was determined by counting blue cells under a Nikon Eclipse TS100 model light microscope, and randomly selected fields were photographed at 1000× magnification, using an Idea SPOT digital camera.

ROS Measurements
Intracellular ROS was determined using the CM-H2DCFDA probe (Molecular Probes, Invitrogen, Carlsbad, CA, USA) and flow cytometry.When oxidized, this probe can be detected by fluorescence with excitation at 485 nm and emission at 515 nm.Irradiated CAFs were seeded in 6 well cell culture plates (1 × 10 5 ) with 2 ml of complete growth medium and incubated overnight at 37˚C, 5% CO 2 .The next day, the cells were washed 2 times with Hanks Balanced Salt Solution with CaCl 2 and MgCl 2 (HBSS) (GIBCO, Invitrogen, Carlsbad, CA) and incubated for 30 min with 2.5 μM CM-H2DCFDA in HBSS.Subsequently the cells were trypsinized, resuspended in 0.5 ml HBSS and immediately analysed by flow cytometry (FACSAria, BD Biosciences).Ten thousand events were collected and analyzed by BD FACSDiva Version 5.0.2.The numbers presented are the median values of the fluorescent intensity.

RNA Preparation and Quality/Quantity Control
Disruption and homogenization of cells were performed in lysis buffer using the MagNa Lyser Instrument (Roche Applied Science, Germany) and according to the manufacturer's protocol.Subsequently, total RNA was isolated with the MagNa Pure Compact Instrument and the MagNa Pure Compact RNA Isolation Kit (Roche Applied Science, Germany) as previously described [17].RNA was quantified by measuring absorbance at 260 nm, and RNA purity was determined by the ratios OD260 nm/280 nm and OD230/280 nm using the NanoDrop instrument (NanoDrop ® ND-1000, Wilmington, USA).
The RNA integrity was determined by electrophoresis using the BioRad Experion Bioanalyzer (Data not shown).All the RNA preparations were verified for possible genomic DNA contamination by a minus-RT-PCR conducted directly on RNA samples, with human ge-nomic DNA as a positive control and amplification of the human housekeeping gene cyclophilin A. Genomic DNA was not detected in RNA preparations (Data not shown).

Probe Generation and Array Hybridization
Prior to probe generation 1 g total RNA was amplified with the Ambion ® MessageAmp TM II aRNA Amplification Kit (Applied Biosystems Inc., USA).RNA samples were transcribed to cDNA using the Invitrogen superscript cDNA synthesis kit (Invitrogen TM , USA).cDNA and Cy3-labelled with the NimbleGen One-Color DNA Labeling Kit and according to the manufacturer's protocol (Roche NimbleGen, Germany).Labelled probes were hybridized to a NimbleGen oligo arrays (12 × 135K) with the Roche NimbleGen hybridization system for 16 hours at 42˚C.The microarray slides were washed according to the manufacturer's protocol and scanned with an Axon GenePix 4000B scanner (Molecular Devices Inc., USA).Features were extracted by using the Nim-bleScan v2.5 software.

Data Analysis and Statistical Analysis
Normalization was carried out using quantile normalization [18].For the purpose of finding differentially expressed genes, we applied an empirical Bayes analysis using the LIMMA package [19], and significance was determined at the 0.05 level corrected for false discovery rate (FDR) using the Benjamini-Hochberg method [20].Principal component analysis (PCA) was carried out on the data in order to visualize the data structure and look for potential outlier samples [21].The differentially expressed genes were annotated by Protein Analysis THrough Evolutionary Relationships (PANTHER; http://www.panther.org/).Gene SetEnrichment Analysis (GSEA) was performed using the R statistical package (http://www.broad.mit.edu/gsea/).The microarray data were prepared according to minimum information about a microarray experiment (MIAME) recommendations and deposited in the Gene Expression Omnibus (GEO) database: http://www.ncbi.nlm.nih.gov/geo/.The GEO accession number for the series is GSE37318.

Survival Rates and CAFs Responses to Ablative Radiation Doses
All the data generated in this study come from cell cultures directly prepared from lung tumor specimens obtained from 8 different donors after surgical resection (Figure 1).In a previous study from our group we describe in detail the isolation and characterization of the tumor fibroblast cultures that have been used also for this study [14].Of note, flow cytometry analyses using the fibroblast specific marker α-SMA reveal a purity above 99.5%, which gives full credibility to the data generated with this material.The intrinsic radioresistance of CAFs in dose escalating trial experiments have also been evaluated and described by us previously [14].A single dose of 18 Gy happens to be sub-lethal for CAFs, however we have observed that single doses above 12 Gy induce enduring DNA damage responses (nuclear foci of 53BP1) and the cells enter into a permanent cell growth arrest or sensecence.Hence, based on our previous experience we have chosen the use of a single dose of 18 Gy as the basis protocol for conducting this study.In previous studies done by others on normal fibroblasts, 24 hours post-irradiation seems to be the time point where the largest number of genes is altered [10].We have therefore set up 24 hours as the single and most relevant time point to analyse changes in gene expression.

Radiation-Induced Gene Expression
Large scale molecular responses in CAFs obtained from NSLC patient material to ablative doses of radiation were studied by measuring genome-wide transcript-level changes using human genome survey microarrays as described in Materials and Methods.After pre-processing and normalization, a total of 680 differentially expressed were found in the group of irradiated CAF samples, of which 127 genes were down-and 553 were up-regulated transcripts with p < 0.05 (Supplemental list 1).Of all differentially expressed genes, 601 could be annotated by GO terms for biological processes in Kegg (http://www.genome.jp/kegg/)and are summarized in Supplemental list 2. Principal component analysis (PCA) revealed that the major variability in the data set is caused by the difference between irradiated and non-irradiated CAFs (Data not shown).The number and distribution of 153 differentially expressed genes with fold change greater than 1 and an adjusted p-value less than 0.05 across different comparisons is shown in Figure 2.

Radiation-Induced Pathways after Functional Categorization
Candidate genes which relate to senescence have been extracted from the GenAge database (http://genomics.senescence.info/genes/human.html)(Table 1).In addition, the expression levels of some differentially expressed genes found in this study have been verified by qRT-PCR (Table 2).Differentially expressed genes were annotated with PANTHER (http://www.pantherdb.org/) to different biological processes: cell cycle regulation and DNA repair (Table 3), oxidative stress (Table 4), apoptosis (Table 5), autophagy (Table 6), and cell communication/cell adhesion (Table 7).

Validation of Expression Data by Functional Assays
Gene expression analysis indicated that some p53-dependent pathways and the cell cycle regulation machinery were altered after IR.One end point of such disturbances is the development of stress-induced cellular senescence.
In our previous study we have performed quantitative and qualitative measurements of senescence on CAFs after exposure to either high single dose or fractionated        irradiation [14].Our data show a potent and permanent induction of cell senescence after a single insult of 18 Gy and Figure 4 illustrates the acquisition of the senescent phenotype by CAFs after exposure to 18 Gy.One of the most drastically altered pathways after irradiation corresponded to DNA damage responses and chromatin rearrangements (Table 3).Previously we have conducted quantitative dose-dependent measurements of DNA damage and repair induced by IR on CAFs.In that study we show that IR provoke substantial DNA damage even at low radiation doses, however we saw that arising nuclear foci were able to resolve at radiation doses lower than 12 Gy.After a single dose with 18 Gy a strong and permanent DNA damage response on cells was established.An illustration of these observations is presented in Figure 5.
Intracellular levels of reactive oxygen species (ROS) were determined as described in detail in the Materials & Methods.ROS production by CAFs was significantly increased in irradiated CAFs (Figure 6).Elevated ROS production has been found to be predominantly a consequence of altered mitochondrial gene expression (Table 4).Increased expression of cytochrome P450, family 3, subfamily A, polypeptide 7 (CYP3A4) and NADH adrenodoxin oxireductase (FDXR) was confirmed by qPCR (Table 2).
Genes associated with regulation of autophagy and apoptosis were also transformed after IR (Table 5 and Table 6).However, expression of apotopic genes like MDM2 p53 binding protein homolog (MDM2) and tumor protein p53 inducible protein 3 (TP53I3), and expression of ATG16 autophagy related 16-like 2 (ATG16L2) gene was confirmed by qPCR (Table 2).
In pilot experiments, apoptosis and autophagy were measured on CAFs at different points after IR by staining for annexin V on living cells and by using Cyto-ID Autophagy Detection Kit (ENZO life Sciences) respecttively (data not shown).These assays failed to show induction of neither apoptosis nor autophagy in CAFs for up to two weeks after exposure to 18 Gy, indicating that the counter balance between the pro-and anti-signaling mechanisms prevented the ultimate induction of such pathways.In addition, the increased expression of genes involved in cell adhesion, collagen type VII alpha (COL7A1) and stromelysin 3 (MMP11), and the cytokine interleukin 12A (IL2A) has been confirmed accordingly (Table 2).

Discussion
The aim of this study was to examine the impact of high-dose ionizing radiation on activated cancer-associated fibroblasts (CAFs) and to predict the consequences of the induced changes on post-radiation tumorigenesis.The overall study showed that approximately 80 % of all differentially expressed genes were up-regulated and 20 % were down-regulated after treatment (Supplemental list 1).In line with previous reports performed on normal tissue fibroblasts, our data show that a large extent of differentially expressed genes belongs to cellular mechanisms related to cell stress, cell cycle, apoptosis, DNA repair, and other pro-survival pathways.It is expected that ionizing radiation induces severe genotoxic stresses in cells, followed by alterations on the expression of genes involved in biological processes like DNA repair, cell proliferation, cell cycle, apoptosis and p53 pathway as has been also shown in a previous study with primary human skin fibroblasts [9].Cell proliferation and cell cycle checkpoints are some of the pathways profoundly affected by ablative doses of ionizing radiation (AIR) on CAFs.Hence, a list of genes involved in cell cycle regu-lation such as cell division cycle homologs CDC6 and CDC34, proliferating cell nuclear antigen (PCNA) and different cylins are all up-regulated, whereas denticless protein homolog (DTL) and minichromosome maintenance component 10 (MCM10) are down-regulated (Table 3).Secondly, (RPA2), Mdm2 p53 binding protein homolog (MDM2), and damage-specific DNA binding protein 2 (DDB2), were both up-regulated.In line with our observations, an increased expression of cylins and PCNA has been also observed in normal fibroblasts after lower dose ionizing radiation [11].In addition, many histones showed increased expression (Table 3).They do not only play an important role in transcription regulation, but are also important team players in DNA repair mechanisms by being responsible for sustaining chromosomal stability [23].Putting it together, these data suggest that CAFs intend to counteract mitotic catastrophe after irradiation.
In normal fibroblast radiation exposure, as well as administration of genotoxic agents, are able to induce growth arrest, a phenomenon described as cellular senescence [24].Senescence is often accompanied by stimuli like oxidative DNA damage, and a compromised replicative cell cycle [25].Although cellular senescence can be ascribed as a self-regulated tumor suppressor mechanism, thereby protecting the organisms from developing cancer, it has been acknowledged that senescent stromal fibroblasts are able to expedite epithelial tumorigenesis [26].Ionizing radiation of CAFs revealed differential expression of a considerable number of genes that are related to senescence and ageing.Candidate genes listed have been shown to be involved in sensecence at different extents (Table 1).
In our hands, high-dose ionizing radiation results in an increase of reactive oxygen species (ROS) formation in CAFs.The provoked enhanced ROS production is reflected by alterations in mainly mitochondrial-generated ROS rather than activation of the NADPH-oxidase (NOX) system (Figure 3 and Table 4).Of note, the disturbances in the redox homeostasis of irradiated CAFs have the potential to lead to increased tumor growth [27].Previous studies have shown that normal fibroblasts express multiple ROS scavenger genes when irradiated with lower doses [13].Here, enhanced expression was only observed for one ROS scavenger, namely gamma glutamyl transferase 1 (GGT1).
The expression of genes involved in the regulation of apoptosis was also modified after AIR, including both pro-and anti-apoptotic signaling pathways (Table 5).Thus, the transcription of genes involved in caspase-mediated apoptosis (caspase recruitment domain-containing protein (PYCARD) and caspase and RIP adapter with death domain (CRADD) was enhanced, which correlates with observations made in irradiated normal tissue fibroblast [11].However, in our in vitro assays we could not observe apoptotic cells during the first two weeks post irradiation.Taken these observations together, we may conclude that the anti-apoptotic signals had stronger influence on the final outcomes on cells.
Of importance, a number of genes regulated by or connected to the tumor suppressor gene p53 were shown to be up-or down regulated.The observed over-expression of MDM2 can result in excessive inactivation of tumor protein p53 in CAFs, thereby diminishing its tumor suppressor function and therefore affecting tumorigenesis [28] or may act in a p53-independent manner as an oncogene [29].It is hereby noted that nine different MDM2 transcripts showed increased expression levels (see also Supplemental list 1).Their functional and not yet defined biological implications on the tumor microenvironment still needs to be elucidated [30].Increased expression levels for MDM2 variants have been reported to be related to radio-sensitivity in the lung [31], and implies the possibility that CAF to some extent become radio-resistant after treatment [32].On the other hand, up-regulation of the TP53-induced glycolysis and apoptosis regulator (TIGAR) may protect irradiated CAFs against reactive oxigen species and apoptosis induced by TP53/p53.It has been shown that fibroblasts are also able to induce an up-regulation of TIGAR expression in cancer cells, thereby protecting cancer cells against apoptosis and autophagy, as recently described in an autophagic tumor stroma model of cancer [33].
Autophagy, which can be ascribed as both cell survival and cell death mechanism has been shown to be activated in tumor cells after ionizing radiation [34].
Here, enhanced expression of a large number of other genes related to autophagy has been observed (Table 6).A role for autophagy in CAFs has been recently shown to be linked to promotion of tumor cell survival [33].However, the establishment of an autophagic response in CAFs after AIR still needs to be confirmed, and whether such stress-induced changes occurring in CAFs could have an impact on tumor cells also remains to be determined.
Pro-angiogenic signals released by CAFs contribute importantly to the overall sustainability of tumors [35,36].However, the observed decreased expression of the proangiogenic factor angiopoietin 1 ANGPT1, a factor that mediates reciprocal interactions between the endothelium and surrounding matrix [37], might result in suppressed tumor growth.Several anti-angiogenic molecules showed increased expression in CAFs after highdose ionizing radiation which might have a direct impact on tumor growth.The matrix-derived angiogenesis inhibitor thrombospondin type I, domain containing 1 (THSD 1) [38], which is able to confer signals like thrombospondin 1 [39][40][41], and interleukin 12A (IL-12A) [42], both may be able to reduce tumor growth rate.In contrast, several semaphorin receptors which are able to confer both, pro-and anti-angiogenic signals are differentially expressed in irradiated CAFs (Table 7).The observed differential expression of plexins PLXNB1, PLXNB3 and PLXA4B imply a role for CAFs on endothelial cell migration [43].Their role in normal lung and in invasive growth and cell migration in lung cancer has been discussed [44,46].Interestingly, the activation of PLXNB1 and PLXNB3 in irradiated CAFs might inhibit integrin-based adhesion and cell migration on the ECM as has been reported earlier for normal fibroblasts [47].
Together with immune cells, fibroblasts are able to support the survival of the tumor [48,49].Although many genes involved in signaling pathways for growth factors (i.e.EGF/EGFR) have been identified in CAFs [48][49][50][51], a differential expression of these particular growth factors or receptors has not been observed by us.Interestingly, two members of the epidermal growth factor (EGF) family downstream of the EGF receptor pathway have been found with increased expression in irradiated CAFs: the autocrine growth factor and mitogen for fibroblasts amphiregulin (AREG), and epidermal growth factor recaptor pathway substrate 8-related protein 2 (EPS8L2).Increased expression of AREG might have an effect on tumor growth since it has been shown that AREG can inhibit growth and apoptosis in certain aggressive adenoma cell lines in culture [52].However, increased expression of EPS8 has been linked to progression of squamous carcinogenesis and might confer pro-tumorigenic effects on the surrounding tumor tissue [53].It is therefore tempting to speculate that alternative signaling mechanisms may deregulate EGFR signaling during carcinogenesis, such as over-expression or activation of EGF/EGFR pathway intermediates.
CAFs express extracellular matrix (ECM) proteins and matrix regulatory agents, thereby providing a favourable microenvironment for cancer cells which is followed by a stimulation of proliferation of tumors and metastasis [54][55][56].The increased expression of collagen type VII, alpha COL7A1 might sustain the stability of the cellular organization within the tumor microenvironment by contributing to epithelial basement membrane organization and adherence by interacting with ECM proteins.On the other hand, the matrix metalloproteinase 11 (stromelysin 3) (MMP11) is the solely significant differentially exessed matrix metalloproteinase after irradiation.The enhanced expression of MMP11 in CAFs could promote cancer progression by remodeling the ECM and confers anti-apoptotic and anti-necrotic effects on tumor cells [57].Numerous studies have linked increased MMP11 expression in different human cancers with poor disease prognosis [58,59], and have been recently suggested to represent a new prognostic indicator for gastric cancer [60].In addition, the ECM protein Gla protein (MGP) showed decreased expression which can be associated with poor prognosis of lung cancer as has been recently reported for colorectal adearcinoma [61].A potential role for MGP as a prognostic marker for breast cancer has also been implied [62].Increased expression levels of multiple splice variants for neuregulin-1 (NRG1) in CAFs (see also Supplemental list 1) might have an impact on neighboring epithelial cells by inducing their growth and differentiation [63,64].Elevated expression of gasdermin B (GSDMB) imply a regulatory role for irradiated CAFs on the adjacent epithelial cells, as augnted expression of GSDML has been correlated to cancer progression in the gastric epithelium [65].
In conclusion, CAFs are able to survive ablative radiation doses, but such insult provoke prominent cellular damages and consequently numerous genes implicated in DNA repair, cellular stress and cell survival pathways are activated.Our analyses show that CAFs display both proand anti-tumorigenic effects after having received a single high-dose of ionizing radiation.These effects might have an impact on the tumor microenvironment in respect to tumor growth and metastasis.The complexity of the observed differential gene expression patterns and implicative biological effects are interesting future objecttives for further elucidations.Genes annotated by the GO term GO:0000278.Progression through the phases of the mitotic cell cycle, the most common eukaryotic cell cycle, which canonically comprises four successive phases called G1, S, G2, and M and includes replication of the genome and the subsequent segregation of chromosomes into daughter cells.In some variant cell cycles nuclear replication or nuclear division may not be followed by cell division, or G1 and G2 phases may be absent.

CELL_CYCLE_GO_000704 9 152
Genes annotated by the GO term GO:0007049.The progression of biochemical and morphological phases and events that occur in a cell during successive cell replication or nuclear replication events.Canonically, the cell cycle comprises the replication and segregation of genetic material followed by the division of the cell, but in endocycles or syncytial cells nuclear replication or nuclear division may not be followed by cell division.Genes annotated by the GO term GO:0051239.Any process that modulates the frequency, rate or extent of an organismal process, the processes pertinent to the function of an organism above the cellular level; includes the integrated processes of tissues and organs.0.24 0.9 0.2 0.17 0.17 0.97 0.45

GROWTH 26
Genes annotated by the GO term GO:0040007.The increase in size or mass of an entire organism, a part of an organism or a cell.0.21 0.64 0.12 0.07 0.

Figure 1 .
Figure 1.A 200× magnification phase contrast micrograph of three weeks old CAFs in monolayer culture is shown in (A).CAFs are characterized by immunostaining with anti-FAP (fibroblast activating protein), a specific marker of reactive fibroblasts (400× magnification) and as shown in (B).

Figure 2 .
Figure 2. Heatmap of 153 differentially expressed genes in irradiated CAFs with fold change greater than 1.0 and an adjusted p-value less than 0.05 are shown.CAF samples are shown as red bar (top) and non-irradiated CAF samples as blue bar (top).Legend (bottom) shows the fold change [log 2 (radiated) -log 2 (normal)] color coded values in the heatmap.

Figure 3 .
Figure 3. Heatmap of 37 selected genes in irradiated CAF samples are shown as red bar (top) and non-irradiated CAF samples as blue bar (top).Legend (bottom) shows the fold change [log 2 (radiated) -log 2 (normal)] color coded values in the heatmap.

Figure 5 .
Figure 5. Persistent DDR (DNA Damage Response).Cells were immunostained for 53BP1 (red, Alexa546) and nuclei were stained with DAPI (Blue).Homogenous nuclear staining of 53BP1 was observed in non-irradiated cells, whereas numerous nuclear foci were observed 24 hours post-radiation.After five days most cells still demonstrated multiple nuclear foci, with a reduced number of foci when compared to earlier time points.

Figure 6 .
Figure 6.The intracellular ROS production was counted as total H 2 O 2 accumulated in cells 24 hours after IR.H 2 O 2 levels were determined in CAFs from 4 different donors by incubating cells with 2.5 mM CM-H2DCFDA for 30 minutes followed by flow cytometric analyses.Ten thousand events were collected for each measurement and the mean values of the relative fluorescence were registered.Panels in (a) illustrate ROS staining in CAFs from one representative donor.Continuous lines correspond to fluorescence patterns in non-irradiated cells; filled peaks correspond to fluorescence intensity in irradiated cells.In (b), mean fluorescent intensities resulting from FACS analysis from four randomly selected donors is presented.Fluorescence intensity from non-irradiated cultures was averaged and normalized (open bars) whereas the irradiated counterparts are presented with X-fold increase in surface labeling relative to control cells (filled bars).The asterix (*) indicates significant differences (p < 0.05) in mean values when compared to the levels in untreated cells.

Table 2 . Verification of selected expressed genes in irradiated CAFs by qPCR.
* Values are shown as fold-change (FC) representing eight independent microarray experiments and # RT-PCR validations in triplicate ± S.D. of each representative RNA sample.

Table 4 . Differentially expressed genes (p < 0.05) related to oxidative stress in irradiated CAFs.
*Values are shown as fold-change (FC) representing eight independent microarray experiments.

. Annotation of differentailly expressed genes in cancer-associated fibroblasts.
Genes annotated by the GO term GO:0014706.The process whose specific outcome is the progression of a striated muscle over time, from its formation to the mature structure.Striated muscle contain fibers that are divided by transverse bands into striations, and cardiac and skeletal muscle are types of striated muscle.Skeletal muscle myoblasts fuse to form myotubes and eventually multinucleated muscle fibers.The fusion of cardiac cells is very rare and can only form binucleate cells.Genes annotated by the GO term GO:0032446.A process by which one or more moieties of a small protein, such as ubiquitin or a ubiquitin-like protein, are covalently attached to a target protein.Genes annotated by the GO term GO:0048523.Any process that stops, prevents or reduces the frequency, rate or extent of cellular processes, those that are carried out at the cellular level, but are not necessarily restricted to a single cell.For example, cell communication occurs among more than one cell, but occurs at the cellular level.Genes annotated by the GO term GO:0006952.Reactions, triggered in response to the presence of a foreign body or the occurrence of an injury, which result in restriction of damage to the organism attacked or prevention/recovery from the infection caused by the attack.Genes annotated by the GO term GO:0048583.Any process that modulates the frequency, rate or extent of a response to a stimulus.Response to stimulus is a change in state or activity of a cell or an organism (in terms of movement, secretion, enzyme production, gene expression, etc.) as a result of a stimulus.
Genes annotated by the GO term GO:0006979.A change in state or activity of a cell or an organism (in terms of movement, secretion, enzyme production, gene expression, etc.) as a result of oxidative stress, a state often resulting from exposure to high levels of reactive oxygen species, e.g.superoxide anions, hydrogen peroxide (H2O2), and hydroxyl radicals.