Activation of the JNKs/ATM-p53 axis is indispensable for the cytoprotection of dermal fibroblasts exposed to UVB radiation

Abstract

Although UVB radiation is mainly absorbed by the epidermis, ~5-10% of its photons reach and affect the upper part of the dermis. Physiologically relevant UVB doses, able to provoke erythema, induce apoptosis in human dermal fibroblasts in vitro, as well as in the dermis of SKH-1 mice. Given the sparse and even contradictory existing information on the effect of UVB radiation on dermal fibroblasts' viability, aim of this work was to unravel the crucial signaling pathways regulating the survival of UVB-treated human dermal fibroblasts. We found that UVB radiation immediately stimulates the phosphorylation of MAPK family members, as well as Akt, and is genotoxic leading to the delayed ATM-p53 axis activation. Akt phosphorylation after UVB radiation is EGFR-mediated and EGFR inhibition leads to a further decrease of viability, while the Akt activator SC79 rescues fibroblasts to an extent by a mechanism involving Nrf2 activation. The known Nrf2 activator sulforaphane also exerts a partial protective effect, although by acting in a distinct mechanism from SC79. On the other hand, inhibition of JNKs or of the ATM-p53 axis leads to a complete loss of viability after UVB irradiation. Interestingly, JNKs activation is necessary for p53 phosphorylation, while the ATM-p53 pathway is required for the long-term activation of JNKs and Akt, reassuring the protection from UVB. Although UVB radiation results in intense and prolonged increase of intracellular ROS levels, classical anti-oxidants, such as Trolox, are unable to affect Akt, JNKs, or p53 phosphorylation and to reverse the loss of fibroblasts' viability. Collectively, here we provide evidence that the main viability-regulating UVB-triggered biochemical pathways act synergistically towards the protection of human dermal fibroblasts, with EGFR/Akt and Nrf2 serving as auxiliary anti-apoptotic machineries, while JNKs/ATM-p53 activation and interplay being overriding and indispensable for the perpetuation of cellular defense and the maintenance of cell viability.

Fig. 1. Exposure of human skin fibroblasts to UVB radiation leads them to a programmed cell death in vitro and in vivo.

A AG01523c and primary human dermal fibroblasts from an adult donor (HDFs) were irradiated under a UVB lamp for durations corresponding to irradiation doses ranging from 35 to 3100 mJ/cm². After incubation at 37 ^oC for 72 h, cells were detached by trypsinization and red-colored metabollicaly active cells after staining with Neutral Red were measured using a haemocytometer. In addition, AG01523c fibroblasts were irradiated with a UVB dose of 700 mJ/cm² before staining with Neutral Red and cell counting 24, 48, and 72 h post-irradiation. Untreated cells served as the reference sample for the estimation of cell viability. Representative graphs out of at least three similar ones from independent experiments are shown here. Asterisks denote statistically significant differences compared to the untreated control at the respective time-point (p < 0.05, Student’s t test). B Cells exposed to a 700 mJ/cm² UVB irradiation dose were observed under a light or fluorescence [after staining with 2 μg/mL 4 ´,6-diamidino-2-phenylindole (DAPI)] optical microscope for the presence of typical apoptotic features; fixed with 50% (v/v) ice-cold ethanol and stained with PI before cell cycle analysis to reveal the incidence of characteristic sub-G1 peaks; subjected to RNA and protein extraction to assess bcl-2 and bax mRNA levels by RT-qPCR and caspase-3 activation by western blot analysis, respectively. Representative microscopic images, histogram plots, relative gene expression graphs, and western blots of three independent experiments are shown. Arrows in the fluorescence microscopic pictures depict regions of condensed/fragmented chromatin in the DAPI stained nuclei (scale bar = 20 μm). M1 marker designates the sub-G1 peak corresponding to apoptotic cells with fragmented DNA preceding the characteristic G1 and G2/M peaks in the histogram plots of PI-stained cells. Asterisks denote statistically significant differences in the gene expression of UVB-exposed fibroblasts in comparison to the untreated control (p < 0.05, Student’s t test). Staurosporine was used as a known apoptosis inducer leading to caspase-3 cleavage, while western blot analysis against actin served as the loading control. C SKH-1 hairless mice were exposed to 350 mJ/cm² of UVB radiation and the dorsal skin was surgically removed 24 and 48 h post-irradiation for the assessment of cleaved caspase-3 in situ expression in the skin sections after immunohistochemical staining. Pictures shown here are representative of three independent experiments. Scale bar = 50 μm.

Fig. 2. Akt and JNKs are the key players in the cellular response of human dermal fibroblasts towards UVB radiation.

A Cells were irradiated with an irradiation dose of 700 mJ/cm² and further incubated at 37 ^oC for the designated time periods before total protein extraction, SDS-PAGE, protein transfer and western blot analysis with antibodies against the phosphorylated forms of JNKs, p38 MAPK, ERKs, and Akt. Representative blots of three independent experiments are shown here. Western blot analysis for the non-phosphorylated forms of the kinases was performed to verify equal loading. B Cells were pre-treated with the specific MAPKs and Akt inhibitors [SB203580 for p38 MAPK (10 μΜ); PD98059 for ERKs (25 μΜ), SP600125 for JNKs (5 μΜ), and MK2206 for Akt (1 μΜ)] for 1 h before UVB treatment with 700 mJ/cm². Cells were further incubated at 37 ^oC for 72 h before trypsinization, staining with Neutral Red and cell counting (N = 4). p for statistically significant differences in comparison to the respective sample without any inhibitor (Student’s t test) is presented in the graph. C Cell cultures were pre-incubated with 1 μΜ of the Akt inhibitor MK2206 and 5 μΜ of the JNKs inhibitor SP600125 for 1 h before exposure to 700 mJ/cm² of UVB radiation, total protein extraction and western blot analysis for phospho-Akt at Ser473 and phospho-JNKs. The non-phosphorylated forms of the kinases served as loading controls. Western blot analyses were repeated three times and a representative experiment is presented here. D SKH-1 hairless mice were exposed to 140 and 350 mJ/cm² of UVB radiation before sacrifice 1 h post-irradiation and removal of the dorsal skin. Non-irradiated (control) vs. irradiated tissue sections were stained using antibodies against the phosphorylated forms of Akt at Ser473 and of JNKs. Pictures of pAkt and pJNKs in situ expression shown here are representative of three independent experiments. Scale bar = 50 μm.

Fig. 3. UVB radiation exerts a ROS-independent genotoxic effect on human dermal fibroblasts in vitro and in vivo.

A Cells were exposed to 700 mJ/cm² of UVB radiation before protein extraction at the designated time-points post-irradiation and western blot analysis for phopsho-ATM, phospho-p53, and total p53. Western blot analysis for actin was performed to verify equal loading. Experiments were repeated three times and representative blots are presented. B Skin fibroblasts were cultured on glass coverslips before UVB treatment with 700 mJ/cm² and further incubated at 37^oC for 24 h. They were then fixed with 4% (w/v) formaldehyde followed by labeling using an antibody against phospho-H2A.X at Ser139 and a FITC-conjugated IgG. Labeled cells were visualized using a confocal laser scanning microscope. Cells exposed to 10 Gy of ionizing radiation served as the positive control. Representative images of three independent experiments are shown here. Scale bar = 20 μm. C SKH-1 hairless mice were exposed to 140 and 350 mJ/cm² of UVB radiation before sacrifice and removal of dorsal skin 6 h post-irradiation. Non-irradiated (control) vs. irradiated tissue sections were subjected to immunohistochemical staining for the phosphorylated form of H2A.X at Ser139. Pictures shown here are representative of three independent experiments and arrows depict γH2A.X-positive cells in the epidermis and dermis of SKH-1 hairless mouse skin. Scale bar = 50 μm. D Cell cultures were pre-incubated with 1 μΜ of the Akt inhibitor MK2206 and 5 μΜ of the JNKs inhibitor SP600125 for 1 h before exposure to 700 mJ/cm² of UVB radiation, protein extraction and western blot analysis for phopsho-p53 at Ser15. Western blot analysis using an anti-actin antibody was performed for the validation of the equal loading. A representative experiment of three similar ones is depicted here. E Human skin fibroblasts were cultured in 96-well plates. When confluent, cells were pre-incubated with 10 μM 2 ´,7 ´-dichlorfluorescein-diacetate (DCFH-DA) for 1 h at 37 ^oC and then exposed to 140, 350, and 700 mJ/cm² of UVB radiation. Intracellular levels of ROS were measured at the designated time-points post-irradiation by recording fluorescence (excitation wavelength: 485 nm, emission wavelength: 520 nm). ROS production was expressed as a % ratio of the untreated control. F Cells were exposed to an irradiation dose of 700 mJ/cm² and further incubated at 37 ^oC for 2, 4, 6, and 10 h before RNA extraction and RT-qPCR analysis using specific primers for HO-1 and NQO1. The 2^−ΔΔCt method was applied to quantify mRNA expression and glyceraldehyde-3-phosphate dehydrogenase (GAPDH) was used as the reference gene. A representative RT-qPCR analysis performed in triplicates is depicted here. The asterisk denotes statistically significant differences in comparison to the untreated control (p < 0.05, Student’s t-test). G Cells were pre-incubated with 20 μΜ of the known anti-oxidant Trolox before UVB treatment with 700 mJ/cm² and total protein extraction at 1 and 12 h for western blot analysis for phospho-Akt (Ser473)/JNKs and phospho-p53, respectively. Representative images of three independent experiments with similar results are shown. The non-phosphorylated forms of the kinases, as well as actin served as the loading controls. H Cells pre-incubated with 20 μΜ Trolox overnight were treated with 10 μM DCFH-DA for 1 h at 37 ^oC before measurement of intracellular ROS levels. ROS production was expressed as a % ratio of the untreated control (N = 3). In parallel, cells pre-treated with 20 μΜ Trolox overnight were exposed to 700 mJ/cm² of UVB radiation. Cultures were further incubated at 37 ^oC for 72 h before trypsinization and measurement of cell number after staining with Neutral Red using a haemocytometer (N = 5, conducted in duplicates). Differences in comparison to the untreated control are considered statistically significant when p < 0.05 (Student’s t test).

Fig. 4. UVB-induced cellular responses are EGFR-Akt mediated in human skin fibroblasts.

A Human dermal fibroblasts were pre-incubated with 100 μM suramin for 1 h before exposure to 700 mJ/cm² of UVB radiation. Then, cells were further incubated at 37 ^oC for 72 h and cell number was counted after staining with Neutral Red. A representative experiment out of three similar ones is shown. p for statistically significant differences in comparison to the respective samples in the absence of suramin (Student’s t test) is presented in the graph. B Cells were pre-treated with 100 μM suramin for 1 h before exposure to 700 mJ/cm² of UVB radiation. Protein extracts after 1 and 12 h were subjected to western blot analysis for phospho-Akt (Ser473)/JNKs and phospho-p53, respectively. Representative blots of three independent experiments with similar results are depicted here. The non-phosphorylated forms of the kinases and actin were analyzed to verify equal loading. C A 100 μΜ suramin solution was analyzed spectrophotometrically in a range from 200 to 600 nm in order to assess its absorption spectrum. PBS served as the negative control. Gray shading marks the wavelength range corresponding to the UVB band (280–315 nm). D Cells were pre-incubated for 1 h with the growth factor receptor inhibitors [STI571 for PDGFR (2 μΜ); SU5402 for FGFR and VEGFR (20 μΜ); SB431542 for TGFβR1 (10 μΜ); I-OMe-AG538 for IGFIR (12 μΜ); AG1478 for EGFR (10 μΜ)] and then exposed to 700 mJ/cm² UVB radiation. Cultures were further incubated at 37 ^oC for 72 h, cells were detached by trypsinization, stained with Neutral Red and counted in a haemocytometer. p for statistically significant differences in comparison to the respective samples with no inhibitor (Student’s t test) is provided. E Cells were incubated with 10 μΜ of the EGFR inhibitor AG1478 for 1 h, exposed to 700 mJ/cm² of UVB radiation and further incubated at 37 ^oC for 1 or 12 h before protein extraction and western blot analysis for phospho-Akt (Ser473)/JNKs or phospho-p53, respectively. Representative blots of three independent experiments are shown here, while total Akt and JNKs, as well as actin were analyzed to validate equal loading. F Cell cultures were treated with 100 ng/ml EGF, exposed to 700 mJ/cm² of UVB radiation and further incubated at 37 ^oC for 72 h. Neutral Red-positive cells were counted in a haemocytometer. A representative graph from three independent similar experiments is presented. p for statistically significant differences in comparison to the respective samples in the absence of the growth factor (Student’s t test) is shown.

Fig. 5. The known Akt activator SC79 protects human dermal fibroblasts from the cytotoxic effect of UVB radiation.

A Cells were pre-treated with 25 μΜ SC79 for 1 h before their exposure to 700 mJ/cm² UVB. After a further incubation at 37 ^oC for 72 h, cells were observed under the microscope or detached by trypsinization and counted after staining with Neutral Red. In addition, RT-qPCR analysis for bcl-2 and bax expression was performed in extracted RNA. Representative microscopic pictures are shown, while numerical values corresponding to the columns of the graphs are means ± standard deviations of three independent experiments conducted in triplicates. p for statistically significant differences in comparison to the respective sample without SC79 (Student’s t test) is presented. In the graphs of gene expression analysis, asterisks denote statistically significant differences in comparison to the respective sample without SC79 (p < 0.05, Student’s t test). B Cells were treated with SC79 at different concentrations (0, 1, 5, 10, and 25 μΜ) before total protein extraction at 1 h post-treatment and western blot analysis for the phosphorylated and non-phosphorylated forms of Akt and JNKs. Representative blots from three similar experiments are shown. C Cells were pre-treated with 1 μΜ of the Akt inhibitor MK2206 and 5 μM of the JNKs inhibitor SP600125 for 1 h or transfected with 50 nM scramble and SignalSilence^® Akt siRNA I sequences, before adding 25 μΜ SC79 and treating the cells with 700 mJ/cm² of UVB radiation. Cell number was counted after 72 h of further incubation at 37 ^oC and staining with Neutral Red. p for statistically significant differences in comparison to SC79/UVB-treated cells in the absence of inhibitor or Akt siRNA (Student’s t test) is presented.

Fig. 6. The Akt activator SC79 and the Nrf2 inducer sulforaphane are photoprotective for human dermal fibroblasts through Nrf2 activation, but without the implication of ROS scavenging.

A Skin fibroblasts were cultured on glass coverslips before treatment with 25 μM of SC79 and they were then fixed with 4% (w/v) formaldehyde in PBS. Labeling was performed using an antibody against Nrf2 and a FITC-conjugated IgG. Labeled cells were visualized using a confocal laser scanning microscope (Scale bar = 20 μm). In addition, RNA was extracted and RT-qPCR analysis was performed to estimate ho-1 and nqo1 mRNA levels. Representative microscopic pictures and representative graphs of means ± standard deviations from three independent experiments conducted in triplicates are depicted here. Asterisks show statistically significant differences compared to the untreated control (p < 0.05, Student’s t-test). B Cells were transfected with 50 nM of either a scramble or an siRNA sequence targeting Nrf2, treated with 25 μΜ of SC79 for 1 h and then exposed to 700 mJ/cm². Cell number was measured after 72 h in Neutral Red-stained cells. A representative experiment is presented out of three independent experiments. p for statistically significant difference in comparison to scramble SC79/UVB-treated cells (Student’s t-test) is provided. C Dermal fibroblasts were transfected with 50 nM of either scramble or Nrf2 siRNA, treated with 25 μΜ of SC79 for 1 h and then exposed to 700 mJ/cm² UVB. Total protein extraction was performed 1 h post-irradiation followed by western blot analysis for phospho-Akt (Ser473) and phospho-JNKs. Representative blots of three independent experiments are shown. Western blot analysis of the non-phosphorylated forms of the kinases was done to evidence equal loading. D Cells were pre-treated with 1 μΜ of the Akt inhibitor MK2206 and 5 μM of the JNKs inhibitor SP600125 for 1 h or transfected with 50 nM scramble and SignalSilence^® Akt siRNA I sequences, before the addition of 25 μΜ SC79 for another 1 h. RT-qPCR analysis for ho-1 gene expression was performed in the extracted RNA samples. A representative experiment from three similar ones is presented. Asterisks denote statistically significant differences in comparison to the respective samples without any inhibitor or siRNA (p < 0.05, Student’s t test). E Cells were treated with 25 μM SC79 or 10 μΜ sulforaphane (SFN) for 1 h and then exposed to an irradiation dose of 700 mJ/cm² UVB before protein extraction and western blot analysis for phospho-Akt (Ser473) and phospho-JNKs, as well as their non-phosphorylated forms. Representative blots of three independent experiments are shown. F Cells were transfected with 50 nM of a scramble or a Nrf2 siRNA sequence, treated with 10 μΜ of SFN for 1 h and then exposed to 700 mJ/cm². Cell number was measured after 72 h in Neutral Red-stained cells. A representative experiment out of three similar ones is presented. p for statistically significant difference in comparison to scramble SFN/UVB-treated cells (Student’s t test) is shown. G Cells were cultured in 96-well plates and treated with 20 μΜ Trolox, 25 μΜ SC79, or 10 μΜ SFN overnight. They were then incubated with 10 μM 2 ´,7 ´-dichlorfluorescein-diacetate (DCFH-DA) for 1 h at 37 ^oC before their exposure to 700 mJ/cm² of UVB radiation. Intracellular levels of ROS were estimated by recording fluorescence (excitation wavelength: 485 nm, emission wavelength: 520 nm). ROS production was expressed as a % ratio of the untreated control. Experiment was repeated three independent times and one representative graph is depicted. p for statistically significant differences compared to the respective sample without Trolox, SC79 or SFN (Student’s t test) is presented. H Cells were treated with 25 μΜ SC79, 10 μΜ SFN or both for 1 h and then exposed to 700 mJ/cm². Cell number was measured after 72 h in Neutral Red-stained cells. A representative graph of three independent experiments is shown here. p for statistically significant difference in comparison to the UVB treated cells in the presence of SC79 or SFN alone (Student’s t test) is presented.

Fig. 7. A functional DNA damage response is indispensable for the activation of human dermal fibroblasts’ defensive mechanisms towards UVB-induced cytotoxicity.

A Cells were pre-incubated with 5 μM of the specific ATM inhibitor KU55933 for 1 h, exposed to 700 mJ/cm² of UVB radiation and further incubated at 37 ^oC for the designated time periods before total protein extraction and western blot analysis using antibodies against phospho-ATM at Ser1981, phospho-Akt at Ser473 and phospho-JNKs. Additionally, cells were transfected with 50 nM of either a predesigned scramble or the human p53 siRNA sequence before their exposure to 700 mJ/cm² of UVB radiation, protein extraction 12 h post-irradiation and western blot analysis for phospho-Akt at Ser473 and phospho-JNKs. The non-phosphorylated forms of Akt and JNKs, as well as actin were analyzed in order to verify equal loading. Representative blots of three independent experiments are shown. Cells were pre-incubated with 5 μM of the specific ATM inhibitor KU55933 for 1 h (B) or transfected with 50 nM of either scramble or p53 siRNA (C), exposed to 700 mJ/cm² of UVB radiation and further incubated at 37 ^oC for 72 h before staining with Neutral Red and cell counting. Graphs are representative of three independent experiments and p for statistically significant differences in comparison to the respective sample without the inhibitor (B) or to the respective scramble (C) is presented. D MDAH041 fibroblasts from a Li-Fraumeni patient were irradiated with UVB irradiation doses ranging from 35 to 3100 mJ/cm². After incubation at 37 ^oC for 72 h, cells were detached by trypsinization, stained with Neutral Red, and measured using a haemocytometer. Untreated cells served as the reference sample for the estimation of cell viability. A representative graph out of three similar ones from independent experiments is shown here. Asterisks denote statistically significant differences in comparison to the respective untreated control (p < 0.05, Student’s t test).

Fig. 8. Proposed model of the UVB-triggered biochemical pathways in human dermal fibroblasts.

A UVB-induced inherent cellular responses. UVB radiation leads human dermal fibroblasts to apoptosis in vitro and in vivo. UVB treatment induces in parallel the EGFR-Akt signaling pathway and the JNKs MAPKs which intervene with the phosphorylation of p53, part of the ROS-independent DNA damage response of the cells towards UVB genotoxicity. EGFR-Akt, JNKs, and ATM-p53 pathways seem to co-operate for the protection towards UVB radiation and the retention of a sufficient percentage of viable cells in the exposed population. This is evidenced by the increased loss of viability after Akt inhibition or siRNA-mediated knocking-down and the unmitigated cell death when directly (or indirectly through JNKs inhibition) impeding the activation of the ATM-p53 pathway. B Interplay of exogenously supplied photoprotective molecules with the intrinsic UVB-induced machinery to amplify the resistance of human skin fibroblasts against UVB radiation. Whereas the known anti-oxidant Trolox does not increase cell survival rate after UVB exposure, extra protection is conferred by other exogenously supplied molecules: suramin acts as a UV shield; EGF activates EGFR; SC79 enhances inherent cellular response (i.e., it activates further Akt and JNKs) and triggers Nrf2 activation downstream of Akt; sulforaphane induces Nrf2 activation independent of the Akt or JNKs signaling pathways. Nrf2-mediated protection by SC79 and sulforaphane is not based on ROS-scavenging and can be attributed to the transcription factor’s ability to safeguard genome integrity or to produce a protective extracellular matrix.