This article is mentioned in:

Introduction

Adjuvant radiotherapy is an effective treatment for thoracic malignancies and has resulted in a significant improvement in the chances of cancer patient's survival over the past decades (1–4). However, while high-energy radiotherapy treatment successfully kills cancer cells, unavoidable radiation exposure to the heart and large arteries occurs during treatment, resulting in radiation-induced secondary cardiovascular disease (CVD) in cancer survivors (5–16). Indeed, epidemiological, clinical and experimental data have established the link between radiation exposure at high and medium doses that are received during radiotherapy (>0.5 Gy) and the risk for CVD (6,9,10,17–20). In addition, recent data suggest that radiation-induced CVD occurs at much lower doses than previously thought (<0.5 Gy) (7,10,19,21–24). However, the underlying cellular and molecular mechanisms of radiation-induced CVD are not fully understood, possibly resulting in improper radiation protection. Most of the clinical and experimental data focus on identifying the late cardiovascular response after radiation exposure, although physiological changes and cellular and molecular damage in the cardiovascular system may happen directly after radiation exposure (7,18,25–32). Moreover, there are currently no validated biomarkers that predict the risk to develop CVD after radiotherapy. Therefore, experimental animal studies of acute and early cardiovascular response after thoracic irradiation may improve our knowledge on early asymptomatic changes in the cardiovascular system after ionizing exposure. This may help in identifying biomarkers indicating developing cardiovascular complications after radiotherapy, which may help in screening patients at risk for developing CVD, thus countermeasures and early medical intervention might be applied to prevent further cardiac toxicity.

Atherosclerosis, a progressive inflammatory disease of the arterial wall, is the main underlying cause of major CVD. Multiple inflammatory markers have been correlated with the pathogenesis of atherosclerosis (33–36). Radiation exposure is also known to trigger the release of several inflammatory and adhesion molecules, involved in the pathogenesis of atherosclerosis (18,26,37–42). Here, we explored a large panel of inflammatory markers in the serum of atherosclerosis-prone ApoE−/− mice at 24 h (acute response) and 1 month (early term response) of thoracic X-ray irradiation. The inflammatory markers were further verified in vitro in coronary artery and microvascular endothelial cell lines exposed to low and high dose X-rays. Serum triglyceride and cholesterol levels, which are known to be strongly associated with atherogenesis (43,44), were also evaluated at 24 h and 1 month post-irradiation. We identified two inflammatory markers, growth differentiation factor-15 (GDF-15) and C-X-C motif chemokine ligand 10 (CXCL10), which were significantly elevated 24 h after irradiation in both female and male mice, which were found to be also increased in irradiated coronary artery and microvascular endothelial cells. Interestingly, we observed gender-specific responses in triglyceride and cholesterol levels 1 month post-irradiation, and in the assessed inflammatory markers at 24 h and 1 month post-irradiation.

Materials and methods

Animals

Animal experiments were approved by the Ethical Committee Animal Studies of the Faculty of Medicine and Health Sciences, Ghent University (ECD 17/60), and were performed in compliance with the Belgian laboratory animal legislation and the European Communities Council Directive of 22 September 2010 (2010/63/EU). Apolipoprotein E-deficient ApoE−/−) mice were purchased from Charles River Laboratories, and animals were housed at the animal facility of UGent. Pups were weaned at the age of three weeks and housed in temperature-controlled, individually ventilated cages, with a 12 h light-dark cycle. They received a standardized mouse chow diet (3.7% fat) and water ad libitum. At the age of 10–12 weeks, ApoE−/− mice were randomly allocated to receive irradiation or sham treatment. In total 72 female and male ApoE−/− mice were used (6 mice per group, except for the control group at 1 month, which were 3 for male and 9 for female mice).

Mouse irradiation

The Small Animal Radiation Research Platform (SARRP; Xstrahl®; in collaboration with INFINITY lab, UGent) was used to irradiate ApoE−/− mice. Mice were anesthetized with isoflurane (5% induction and 2% maintenance) and subjected to a full body CT-scan (50 kV, 1.5 mA, 360 projections >360°, 1 mm aluminium filter) prior to irradiation. CT images were analyzed by using Muriplan software (Xstrahl®) to determine the coordinates of the isocenter for subsequent irradiation. Thoracic irradiation was performed in a ventro-dorsal direction, with 220-kV X-rays, operating at 13 mA and filtered with 0.15 mm of copper, operating with a dose rate of 3.4 Gy/min. The field size, created by the collimator of 10×10 mm, was encompassing both carotid arteries, the aortic arch and apical portion of the heart. The doses that were delivered to the heart region are 0 Gy (sham treatment; also referred to as ‘controls’), 0.1 and 10 Gy.

Cell culture

Telomerase Immortalized human Microvascular Endothelial cells (TIME) from the American Type Cell Culture (ATCC, France; ATCC® CRL-4025™) (https://www.lgcstandards-atcc.org/Products/All/CRL-4025) was used. In addition, TICAE cells, which are primary human coronary artery endothelial cells from the European Collection of Authenticated Cell Cultures (ECACC; HCAECs cat. no. 300-05a) that were transduced with retroviruses bearing the est2 gene, a yeast homologue of the human TERT protein (45,46), were used. TIME and TICAE cells are from a male human donor. MesoEndo Cell Growth Medium (Sigma-Aldrich; Merck KGaA) was used for TIME and TICAE cells. The passage number that was used in the experiments is 38 for TIME cells and 33 for TICAE cells. Cells were grown at 37°C in a humidified incubator supplemented with 5% CO2 and were split with a 0.05% trypsin supplemented with 0.02% ethylenediaminetetraacetic acid (EDTA) every 3–4 days. Moxi Z Mini Automated Cell Counter (Orflo Technologies) was used to count the cells. For the cytokine detection experiment, TIME and TICAE cells were seeded in 6-well plate at a density of 2.5×105 per well in 6 biological replicates. Three days later, cells reached 100% confluence. The medium was changed before irradiation, and at 24 and 72 h after irradiation the supernatant was collected.

Cell irradiation

Both TIME and TICAE cells were irradiated at 100% confluence. Single X-rays doses [0 Gy (also referred to as ‘control’), 0.1 and 5 Gy] were applied to the cells using a vertical X-ray beam using a Xstrahl RX generator (320 kV, Filtration: 3.8 mm Al and 1 mm Cu, tube current: 12 mA) (Camberley), at a dose rate of 0.5 Gy/min. X-irradiation was performed in accordance to ISO 4037 and under ISO 17025 accreditation of the Laboratory for Nuclear Calibrations (LNK) of the Belgian Nuclear Research Centre (SCK•CEN). Cells were moved to the irradiation facility using a mobile incubator.

Blood sampling

The blood collection from each mouse was done at two different time points: 1) One week before irradiation and 2) a specific time period after radiation exposure (24 h or 1 month). The mice were anesthetized intraperitoneally with xylazine (10 mg/kg) and ketamine (80 mg/kg) using a 30G needle, and anesthetic depth was assessed using the pedal withdrawal reflex. Retro-orbital puncture was performed to take a 100 µl blood sample with a Brand micropipette one week before irradiation, and at the specific time points after irradiation, where the eye was pulled out to collect at least 500 µl blood. Immediately after the second blood sampling, anesthetized mice were euthanized due to exsanguination via transcardial perfusion. To obtain serum, the collected blood was allowed to clot in eppendorf tubes for 1.5 h at room temperature and was subsequently centrifuged at 1,500 g for 15 min at 4°C. The supernatant was removed and centrifuged again at the same conditions to obtain pure serum. The latter was collected in eppendorf tubes and stored at −80°C.

Triglyceride quantification

The Triglyceride Quantification assay kit (Abcam, ab65336) was used according to the manufacturer's instructions. Briefly, a 50 µl triglyceride standard curve was prepared and 1 µl serum was added to 49 µl assay buffer and all samples were performed in duplicate. Subsequently, 2 µl of lipase, which converts the triglycerides to free fatty acids and glycerol, was added to each well and incubated for 20 min at room temperature. Then, 50 µl reaction mix consisting of 46 µl triglyceride assay buffer, 2 µl triglyceride probe and 2 µl triglyceride enzyme mix, was added to each well and incubated at room temperature for 60 min protected from light. The glycerol molecules are oxidized to generate hydrogen peroxide (H2O2), which reacts with the probe to generate fluorescence that was measured at Ex/Em 535/587 nm with a microplate reader (Victor3, 1420 multilabel counter, PerkinElmer).

Cholesterol quantification

The Cholesterol Assay kit (Abcam, ab65390) was used according to the manufacturer's instructions. A 50 µl cholesterol standard curve was prepared. The serum samples were diluted 1/400 in assay buffer and 50 µl of this diluted serum was used. Then, 50 µl of a cholesterol reaction mix, consisting of 44 µl assay buffer, 2 µl cholesterol probe, 2 µl enzyme mix, and 2 µl cholesterol esterase was added to each well for an incubation period of 60 min at 37°C. The cholesterol esterase hydrolyzes cholesteryl esters to free cholesterol. Subsequently, cholesterol oxidase specifically recognizes free cholesterol and produce H2O2 that reacts with the probe to generate fluorescence (Ex/Em=538/587 nm) that was measured with a microplate reader (Victor3, 1420 multilabel counter, PerkinElmer).

Inflammatory cytokine detection

Simultaneous detection of multiple inflammatory markers in mice serum samples (Adiponectin, CRP, CXCL10, Endoglin, FGF-basic, GDF-15, ICAM-1, IL-6, IL-1β, MCP-1, P-selectin, PAI-1, PCSK9, and uPAR) or in cell supernatant samples (GDF-15, CXCL10, ICAM-1, MCP-1, uPAR, PAI-1, P-selectin, FGF-basic and IL-6) was performed using multiplex bead assay (Luminex® MAGPIX Assay, R&D Systems) following manufacturer's instructions. Luminex® MAGPIX technology is based on the use of a mixture of magnetic microspheres that is added to the sample in a pre-coated plate with analyte-specific capture antibodies, followed by adding specific biotinylated detection antibodies and phycoerythrin that binds to the biotinylated antibodies. Magnetic beads are captured and held in a monolayer by a magnet, while the beads are illuminated by two spectrally distinct light-emitting diodes (LEDs), one to identify the analyte and the other to determine amount of analyte based on the magnitude of the phycoerythrin-derived signal. Signals are captured and imaged with a charge-coupled device (CDD) camera (47–50). Briefly, standards, serum or supernatant samples, and magnetic microparticles cocktail were incubated into cytokine-specific antibodies in a pre-coated 96-well plate. After 2-h incubation in the dark, the plate was washed, and a biotinylated detection antibody cocktail was added. After applying a second wash to remove the unbound biotinylated antibodies, a streptavidin-phycoerythrin conjugate was added to each well to bind to the biotinylated antibodies. Further, a final wash was performed and the microparticles were detected using the Luminex® MAGPIX Analyzer. Inflammatory markers concentration in the supernatant of TIME and TICAE cells was normalized to cell count performed using IncuCyte ZOOM™ phase contrast imaging with ×10 magnification.

Statistical analysis

Analysis of cholesterol, triglyceride and supernatant inflammatory cytokine data were performed using Kruskal-Wallis test followed by Dunn's test, and the P-value was adjusted using Benjamini-Hochberg procedure to control the False Discovery Rate. Data are presented as mean ± standard error of the mean. GraphPad Prism 5.01 (GraphPad Software Inc.) was used for these statistical analysis. For in vivo serum inflammatory cytokines, the fold change in inflammatory marker levels before and after irradiation was first calculated, followed by statistical analysis. Statistical analysis was done with a Kruskal-Wallis test followed by Dunn's test and the P-value was adjusted using Benjamini-Hochberg procedure, using ggsignif R package. The results are considered significant if the P-value is <0.05. Data are presented as boxplot, ggplot2 R package, showing the median with the lower and upper hinges that correspond to the first and third quartiles (the 25 and 75th percentiles) and upper/lower whisker extends from the hinges to the largest/smallest value no further than 1.5 * IQR from the corresponding upper and lower hinge, respectively. Data beyond the end of the whiskers are called ‘outlying’ points and are plotted individually. In vivo data are represented separately in female ApoE−/− mice and male ApoE−/− mice, or as the sum of female and male data, which are referred to as combined data.

Results

Thoracic irradiation induces early term alterations in triglyceride and cholesterol levels in ApoE−/− mice

At 24 h post-thoracic irradiation

Thoracic irradiation of female ApoE−/− mice with 10 Gy induced a significant decrease in triglyceride and cholesterol levels at 24 h after irradiation, compared to age-matched controls (Fig. 1A and B), (summarized in Table SI). In male ApoE−/− mice, thoracic irradiation induced a significant increase in triglyceride level only at 0.1 Gy, but no significant changes in cholesterol level was observed at 24 h after irradiation (Fig. 1A and B). Female and male ApoE−/− mice combined data showed no significant changes in triglyceride and cholesterol levels at 24 h after irradiation (Fig. 1A and B).

Figure 1. Triglyceride and total cholesterol concentration in serum of female (left panels), male (middle panels) and combined female and male ApoE−/− mice (right panels). (A) Triglyceride level at 24 h, (B) total cholesterol at 24 h, (C) triglyceride level at 1 month and (D) total cholesterol at 1 month after irradiation. Values represent average ± SEM of 6 mice per group, except for the 1 month time point in the 0 Gy male group, which had 3 mice. *P<0.05 and **P<0.01 vs. age-matched 0 Gy controls. p.i., post irradiation.

At 1 month post thoracic irradiation

At 1 month after irradiation, a significant increase in triglyceride levels in female ApoE−/− mice was observed at 10 Gy, whereas total cholesterol levels were significantly decreased at 0.1 and 10 Gy, compared to age-matched controls (Fig. 1C and D). In male ApoE−/− mice, a significant decrease in triglyceride level was observed in a dose-dependent manner at 1 month after irradiation, but no significant changes in cholesterol level was observed (Fig. 1C and D). Female and male ApoE−/− mice combined data showed no significant changes in triglyceride levels, while a significant dose-dependent decrease in cholesterol level at 1 month after irradiation was observed (Fig. 1C and D).

Thoracic irradiation induces acute and early term systemic inflammatory response in ApoE−/− mice

To assess the inflammatory response, fourteen different cytokines, which are involved in the pathogenesis of atherosclerosis (Adiponectin, CRP, CXCL10, Endoglin, FGF-basic, GDF-15, IGMA-1, IL-6, IL-1β, MCP-1, P-selectin, PAI-1, PCSK9, and uPAR), have been assessed in serum samples of female and male ApoE−/− mice at 24 h and 1 month after local thoracic X-ray irradiation. Fold changes between the baseline cytokines levels and at 24 h or at 1 month post irradiation cytokine levels were analyzed (summarized in Table SII).

At 24 h post thoracic irradiation

At 24 h post thoracic irradiation, a significant increase in GDF-15 at 0.1 and 10 Gy in female ApoE−/− mice and a significant increase at 10 Gy in male ApoE−/− mice were observed (Fig. 2A). Female and male ApoE−/− mice combined data showed a significant increase in GDF-15 at 10 Gy. A significant increase in CXCL10 was also observed at 10 Gy in female and male ApoE−/− mice, as well as in the combined female and male data (Fig. 2B). In addition, a significant increase in ICAM-1 was observed at 10 Gy only in male ApoE−/− mice (Fig. 2C). Whereas, thoracic irradiation induced a significant decrease in MCP-1 at 0.1 and 10 Gy only in female ApoE−/− mice (Fig. 2D). Moreover, a significant decrease in female uPAR at 10 Gy was observed, and female and male uPAR combined data showed a significant decrease at 0.1 and 10 Gy (Fig. 2E). Next to that, in male ApoE−/− mice, there was a significant increase in PCSK9 at 0.1 Gy (Fig. 2F), and an increase in P-selectin at 10 Gy (Fig. 2G). Nonetheless, no significant changes were observed in adiponectin, PAI-1, CRP, endoglin, FGF-basic, IL-1β and IL-6 at 24 h post thoracic irradiation (Fig. S1).

Figure 2. Systemic inflammatory response in female (left panels), male (middle panels) and combined female and male ApoE−/− mice (right panels) at 24 h post thoracic irradiation. (A) GDF-15, (B) CXCL10, (C) ICAM-1, (D) MCP-1, (E) uPAR, (F) PCSK9 and (G) P-selectin cytokines were analysed. The remaining are listed in Fig. S1. Data are presented as the fold change of the post-irradiation cytokine level relative to the baseline level. Statistical analysis was performed using a Kruskal-Wallis test, with the P-value being adjusted using the Benjamini-Hochberg method. Data are presented as a boxplot showing the median value of 6 mice per group. p.i., post irradiation; GDF-15, growth differentiation factor-15; CXCL10, C-X-C motif chemokine ligand 10; ICAM-1, intercellular adhesion molecule-1; MCP-1, monocyte chemoattractant protein-1; uPAR, urokinase-type plasminogen activator receptor; PCSK9, proprotein convertase subtilisin/kexin type 9.

At 1 month post thoracic irradiation

The systemic inflammatory response at 1 month was different from the acute response at 24 h post thoracic irradiation. Adiponectin was significantly increased at 10 Gy post-irradiation only in male ApoE−/− mice (Fig. 3A). CXCL10 and IL-6 were significantly decreased at 0.1 Gy in female ApoE−/− mice, and in combined female and male data (Fig. 3B and E). FGF-basic level was significantly decreased at 10 Gy in female mice, and in the female and male mice combined data (Fig. 3C). Furthermore, PAI-1 showed a significant increase at 10 Gy in male ApoE−/− mice (Fig. 3D). No significant changes were observed in Endoglin, P-selectin, GDF-15, IL-1β, ICAM-1, MCP-1, PCSK-9, and uPAR at 1 month post thoracic irradiation (Fig. S2).

Figure 3. Systemic inflammatory response in female (left panels), male (middle panels) and combined female and male ApoE−/− mice (right panels) at 1 month post thoracic irradiation. (A) Adiponectin, (B) CXCL10, (C) FGF-basic, (D) PAI-1 and (E) IL-6 cytokines were assessed. The remaining are listed in Fig. S2. Data are presented as the fold change of the post-irradiation cytokine level relative to the baseline inflammatory cytokine level. Data are considered significant when P<0.05. Data are presented as boxplots showing the median of 6 mice per group, except for the male 0 Gy group, which had 3 mice. p.i., post irradiation; m, month; CXCL10, C-X-C motif chemokine ligand 10; FGF-basic, basic fibroblast growth factor; PAI-1, plasminogen activator inhibitor-1.

X-irradiation induces inflammatory response in coronary artery and microvascular endothelial cells

To verify whether the observed inflammatory responses in serum cytokines at 24 h and 1 month after irradiation is linked to endothelial cell responses, two endothelial cell lines (TICAE and TIME cells, originating from a male donor) were irradiated at low dose (0.1 Gy) or high dose (5 Gy), with supernatant being collected at 24 h or 72 h after irradiation. The 72 h time point was chosen since it is not possible to keep cells in culture for 1 month post-irradiation without passaging them. In addition, the high dose was chosen to be 5 Gy in vitro, since we previously reported that 5 Gy induced significant apoptotic cell death from 4 h after X-irradiation in TICAE and TIME cells, which increased with time (51), and also considering that we are irradiating a monolayer of endothelial cells (not tissue like the in vivo situation). The cytokines tested were GDF-15, CXCL10, ICAM-1, MCP-1, uPAR, IL-6, PAI-1, P-selectin, and FGF-basic (summarized in Table SIII).

At 24 h post endothelial irradiation

At 24 h after irradiation, GDF-15 and CXCL10 increased significantly at 5 Gy in TICAE and TIME cells (Fig. 4A, B, J and K) which correspond to the increase in these markers in female and male mice at 24 h post irradiation (Fig. 2A and B). A significant increase in ICAM-1 was observed at 5 Gy in TIME cells (Fig. 4L), which may correlate with the increase in serum ICAM-1 in male mice at 10 Gy for the 24 h time point (Fig. 2C). Additionally, MCP-1 level was not changed in response to radiation in TICAE and TIME cells at 24 h post irradiation (Fig. 4D and M), and this was the case in the irradiated male mice at the same time point (Fig. 2D). Moreover, a significant increase in P-selectin was observed at 5 Gy in TICAE and TIME cells at 24 h post irradiation (Fig. 4H and Q), which corresponds to the observed trend of increase in male mice data (Fig. 2G). The levels of uPAR, IL-6 and PAI-1 were significantly increased in TICAE and TIME cells, and FGF-basic increased in TICAE cells at 5 Gy after 24 h of irradiation (Fig. 4E-G, I, N, O and P), while this was not the case in the irradiated female and male mice at the same time point.

Figure 4. Inflammatory response in endothelial cells at 24 h after irradiation. The response of various inflammatory markers at 24 h after 0.1 and 5 Gy irradiation is represented in (A-I) TICAE cells and (J-R) TIME cells, normalized to cell count. The Kruskal-Wallis test was used to analyse the data and the P-value was adjusted using the Benjamini-Hochberg method. Values represent the average ± SEM of 6 biological replicates. *P<0.05, **P<0.01 and ***P<0.001 vs. 0 Gy. p.i., post irradiation; GDF-15, growth differentiation factor-15; CXCL10, C-X-C motif chemokine ligand 10; ICAM-1, intercellular adhesion molecule-1; MCP-1, monocyte chemoattractant protein-1; uPAR, urokinase-type plasminogen activator receptor; PAI-1, plasminogen activator inhibitor-1; FGF-basic, basic fibroblast growth factor.

At 72 h post endothelial irradiation

At 72 h after irradiation, GDF-15 was still significantly elevated in TICAE and TIME cells at 5 Gy (Fig. 5A and J). CXCL10, ICAM-1, MCP-1, uPAR, IL-6 and FGF-basic were also significantly elevated at 5 Gy in the irradiated TICAE cells (Fig. 5B-E, F and I), while ICAM-1 and MCP-1 significantly decreased in a dose-dependent manner in the irradiated TIME cells at 72 h post irradiation (Fig. 5L and M). Moreover, there was a significant increase in P-selectin at 5 Gy in TICAE cells, and significant increase in PAI-1 at 5 Gy in TICAE and TIME cells (Fig. 5G, H and P).

Figure 5. Inflammatory response in endothelial cells at 72 h after irradiation. The response of various inflammatory markers at 72 h after 0.1 and 5 Gy irradiation is presented in (A-I) TICAE cells and (J-R) TIME cells. Data were normalized to cell count. The Kruskal-Wallis test was used to analyse the data and the P-value was adjusted using the Benjamini-Hochberg method. Values represent the average ± SEM of 6 biological replicates. *P<0.05, **P<0.01 and ***P<0.001 vs. 0 Gy. p.i., post irradiation; GDF-15, growth differentiation factor-15; CXCL10, C-X-C motif chemokine ligand 10; ICAM-1, intercellular adhesion molecule-1; MCP-1, monocyte chemoattractant protein-1; uPAR, urokinase-type plasminogen activator receptor; PAI-1, plasminogen activator inhibitor-1; FGF-basic, basic fibroblast growth factor.

Discussion

Thoracic cancer radiotherapy significantly increases the risk for developing CVD (6,9,10,52,53). The underlying pathophysiology is complex and the mechanisms are not fully understood (6,7,9,17–21). Previous in vivo studies have identified late cardiovascular responses to ionizing radiation exposure, which was mainly manifested by pro-inflammatory responses and accelerated formation of atherosclerotic lesions/plaques 2–15 months after irradiation (17,18,26–29,54). Most of these pre-clinical studies were performed in a single gender, with limited data available related to possible gender differences. Here, we investigated acute (24 h) and early term (1 month) cardiovascular responses to ionizing radiation exposure, by assessing triglyceride and total cholesterol levels, and exploring a large panel of inflammatory markers in serum of female and male ApoE−/− mice that received low or high doses of local thoracic X-ray irradiation. We report, for the first time, that local thoracic irradiation of ApoE−/− mice increases serum GDF-15 and CXCL10 in both female and male mice 24 h after high dose irradiation (10 Gy). GDF-15 and CXCL10 levels were also significantly elevated at 24 h in irradiated coronary artery and microvascular endothelial (TICAE and TIME) cells in vitro after high dose of X-ray irradiation. In addition, we report gender-specific responses in the tested triglyceride and cholesterol levels at 1 month after irradiation, and in the assessed inflammatory markers at 24 h and 1 month post-irradiation. Below we discuss these findings in more detail.

Increased cholesterol and triglyceride levels in atherosclerosis have been shown in several clinical studies (55), and it was suggested that triglyceride level can be used as independent risk factor biomarker for coronary artery disease (56). Increased cholesterol and triglyceride levels may contribute to a large number of biological actions and consequences, including inflicting endothelial cell injury, increasing adhesion molecule expression, increasing leukocytes recruitment, as well as the formation of foam cells, therefore contributing to the atherosclerotic process (43,44). We observed different responses in cholesterol and triglyceride levels in the irradiated female and male ApoE−/− mice, especially at 1 month after exposure. Although triglyceride levels were increased in 10 Gy-irradiated female mice, levels dose-dependently decreased in male mice, demonstrating that triglyceride levels display gender specific responses. In contrast, cholesterol levels dose-dependently decreased in female mice, and showed a decreasing trend in male mice, resulting in a significant dose-dependent decrease when combining female and male responses 1 month post-irradiation. A similar observation for the decrease in total cholesterol and the increase in triglyceride levels in female mice was reported in female breast cancer patients after 50–60 Gy of radiotherapy which were given over 5 weeks (57). Previous studies performed 2–5 months after irradiation of ApoE−/− mice, revealed variable responses in cholesterol and triglyceride levels depending mainly on the time point after exposure, gender used in the experiment, and on radiation doses/quality used (17,18,27,29). Cholesterol and triglyceride changes can perhaps be secondary to the inflammatory or oxidative stress responses after radiation exposure, though further investigations are needed to unveil the involved mechanisms.

Our results further demonstrate that local thoracic irradiation induced an increase in serum GDF-15, which was significant at low and high doses (0.1 and 10 Gy) in female mice and at high dose in male mice 24 h after irradiation. In vitro, GDF-15 was secreted from irradiated TICAE and TIME cells at 24 h after high dose (5 Gy) irradiation. An increased GDF-15 gene expression was previously observed in human aortic endothelial cells at 4–24 h after 4 Gy (58), and in carotid arteries of male ApoE−/− mice at 1 week after 14 Gy of local neck X-ray irradiation (59). GDF-15 is a member of the transforming growth factor β superfamily that increases its expression under inflammatory conditions (60,61). Elevated GDF-15 serum levels have been associated with an increased risk for a range of CVD, including atherosclerosis, and currently being evaluated as a biomarker in CVD (62–65). Previous studies revealed that GDF-15 may contribute to the initiation and the progression of atherosclerotic lesions by regulating apoptosis and IL-6-dependent inflammatory responses (66), promoting migration of macrophage, and by contributing to plaque instability (67). In line with this, our in vitro data showed an increased IL-6 level in TICAE and TIME cells 24 h after 5 Gy irradiation. GDF-15 may play an important role in ionizing radiation-induced endothelial cell senescence through an oxidative stress-mediated p16 pathway (58). Another inflammatory marker of potential interest is CXCL10, also known as Interferon-γ-inducible protein 10 (IP-10), which acts as a chemoattractant cytokine. CXCL10 was shown to promote atherosclerosis by recruitment and retention of activated T lymphocytes to vascular wall lesions during the atherogenesis process (68,69), and also is being investigated as a potential biomarker for CVD (70–73). In our study, the serum level of CXCL10 was significantly elevated in female and male ApoE−/− mice 24 h after high dose thoracic irradiation, and was found to be secreted from both TICAE and TIME cells 24 h post-irradiation. In line with our in vitro data, CXCL10 gene expression was previously reported to be upregulated in human coronary artery endothelial cells after 10 Gy of fractionated X-irradiation (74), and in human umbilical vascular endothelial cells following exposure to 20 Gy of acute gamma irradiation (75).

One of the earliest responses to endothelial cell injury, which is considered one potential initiating event of the atherosclerotic process, is upregulation of adhesion molecules including ICAM-1 and P-selectin, leading to leukocyte adherence to the endothelium (30,76,77). Multiple in vitro and in vivo studies have reported post-irradiation increases in ICAM-1 and P-selectin levels, thereby promoting leucocyte adhesion to the endothelium (26,78–83). However, most of the in vivo studies were performed in endothelial cells of irradiated arteries weeks to months after radiation exposure. Here, we report an acute dose-dependent increase in serum ICAM-1 level 24 h after thoracic irradiation in male ApoE−/− mice, and this increase was stabilized 1 month later. Previous observations have shown that local irradiation of carotid arteries in female and male ApoE−/− mice did not induce ICAM-1 changes at 1 and 22 weeks after 14 Gy X-ray exposure (17,18), indicating that the early 24 h response reported here may be transient in nature. We further found a corresponding ICAM-1 increase in the irradiated TIME cells at 24 h for the 5 Gy dose, which is in line with previous in vitro studies performed at the same time window (39,84). Moreover, we showed that the serum level of P-selectin was increased 24 h after irradiation, again only in male ApoE−/− mice. This elevated P-selectin level was also observed in irradiated TICAE and TIME cells, which confirms previous in vitro findings (79). The inflammatory responses observed at high irradiation dose (5 Gy) in TICAE and TIME cells could be an injury response, since an increased DNA damage and persistent cell death were previously observed in these cells after 5 Gy X-irradiation (51).

Our data further demonstrate a significant decrease in atheroprotective basic fibroblast growth factor (FGF-basic), in female ApoE−/− mice at 10 Gy, and a decreasing trend in male ApoE−/− mice 1 month after irradiation, resulting is a significant dose-dependent decrease in the combined female and male data. A previous study performed in non-Hodgkin lymphoma cancer patients showed a significant decrease in FGF-2 serum level after doses ranging between 6 and 52 Gy of radiotherapy (85), and another study performed in patients with different tumour histotypes, also reported that FGF-2 serum level decreased after radiotherapy (86). FGF-basic was observed to have a protective effect against irradiation (87), since it inhibited radiation-induced apoptosis of endothelial cells under both in vitro and in vivo conditions (88,89), and was also reported to decrease VCAM-1 expression and macrophage presence in atherosclerotic plaques from rabbits fed with a high cholesterol diet (90).

In contrast to the proatherogenic inflammatory response after thoracic irradiation, our results showed an acute decrease in the proatherogenic cytokine MCP-1, only observed in female mice 24 h after irradiation. MCP-1 is involved in atherosclerosis initiation and progression by recruiting monocytes and contributing to macrophage infiltration into the subendothelial cell layer (91,92). Interestingly, it has been shown that post-irradiation MCP-1-mediated chemoattraction is regulated by the proatherogenic uPAR expression (93,94). In line with this observation, our results showed a decrease in uPAR serum level only in female ApoE−/− mice 24 h after 10 Gy exposure. It is worth mentioning that the observed alterations in inflammatory markers, as discussed, are limited to multiplex bead assay assessment, and further validation using other approaches, such as western blotting or RT-qPCR, are required.

The gender difference in response to radiation exposure could be explained by hormone differences between male and female. Several groups have reported that estrogen is protective against vascular dysfunction and atherosclerotic lesions in mice (95–98). Estrogen is known to exert protective and beneficial effects in the cardiovascular system by improving vascular function, increasing NO production and by inhibiting proliferation and migration of vascular smooth muscle cells (99–101). It was also shown that atherosclerotic lesions were significantly less extensive in female ApoE−/− mice than in male mice (101). Indeed, in our experiment, we observed an increase in the pro-inflammatory markers ICAM-1, P-selectin in male mice while a decrease in the pro-inflammatory MCP-1 and uPAR levels in female mice. Though, further investigations are required to unveil mechanisms behind the observed gender specific response after irradiation in ApoE−/− mice, by scrutinizing the effect of sex steroid hormones on cytokine response in vitro, or by using knock-down/out techniques in vivo.

Taken together, our results reveal acute and early term inflammatory responses after X-ray exposure of ApoE−/− mice and of coronary artery and microvascular endothelial cells. Future research is needed to fully grasp the scope of early changes in the cardiovascular system after thoracic irradiation, and to determine whether GDF-15 and CXCL10 could be used as potential biomarkers for early detection of cardiovascular risks in thoracic radiotherapy-treated patients, thus identifying patients who may benefit from early medical intervention.

Supplementary Material

Supporting Data

Acknowledgements

The authors would like to thank Dr Kenneth Raj (Centre for Radiation, Chemical and Environmental Hazards, Public Health England, Didcot, UK) who donated the TICAE cell line.

Funding

This work was supported by the Fund for Scientific Research Flanders, Belgium (grant no. G040720N). RR was supported by a doctoral grant obtained from Belgian Nuclear Research Centre/Ghent University.

Availability of data and materials

The datasets used and/or analysed during the present study are available from the corresponding author on reasonable request.

Authors' contributions

RR conducted ex-vivo and in vitro experiments, guided and supervised the in vivo irradiation and blood collection experiments, and wrote the manuscript. MC and EC conducted in vivo irradiation and blood sampling, as well as the cholesterol and triglyceride experiments. MM analysed the in vivo experiments. RR and AA confirm the authenticity of all the raw data. ED, SB, AA and LL designed the experiments and supervised the study. All authors, except MC, contributed equally in reviewing the manuscript. All authors read and approved the final manuscript.

Ethics approval and consent to participate

Mice were treated according to the European Ethics Committee guidelines and the study protocol was approved by the Animal Experiment Ethical Committee of the Faculty of Medicine and Health Sciences, Ghent University, Belgium (approval no. ECD 17/60).

Patient consent for publication

Not applicable

Competing interests

The authors declare that they have no competing interests.

References