- Email: [email protected]
The radioprotector O-phospho-tyrosine stimulates DNA-repair via epidermal growth factor receptor- and DNA-dependent kinase phosphorylation
Radiotherapy and Oncology 84 (2007) 328–334 www.thegreenjournal.com
Experimental radiobiology
The radioprotector O-phospho-tyrosine stimulates DNA-repair via epidermal growth factor receptor- and DNA-dependent kinase phosphorylation Klaus Dittmanna,*, Claus Mayera, Gabriele Wannera, Rainer Kehlbachb, H. Peter Rodemanna a
Division of Radiobiology and Molecular Environmental Research, Department of Radiation Oncology, and bDepartment of ¨bingen, Tu ¨bingen, Germany Radiology, University of Tu
Abstract Background and purpose: Purpose of the study was to elucidate the underlying molecular mechanism of the radioprotector O-phospho-tyrosine (P-Tyr). Methods: Molecular effects of P-Tyr at the level of EGFR responses were investigated in vitro with bronchial carcinoma cell line A549. Nuclear EGFR transport and DNA-PK activation were quantified after Western blotting. Residual DNAdamages were quantified by help of cH2AX focus assay. Results: As determined by dose–response curves, treatment of cells with P-Tyr for 16 h before irradiation results in radioprotection. Simultaneous treatment with EGFR blocking antibody Cetuximab abolished P-Tyr associated radioprotection. At the molecular level P-Tyr mediated a general phosphorylation of EGFR and a pronounced phosphorylation of nuclear EGFR at residue Thr No. 654, also observed after treatment with ionizing radiation. This phosphorylation was associated with nuclear EGFR accumulation. Moreover, P-Tyr-triggered EGFR nuclear accumulation was associated with phosphorylation of DNA-PK at Thr 2609. This activated form of DNA-PK was not DNA associated, but after radiation, DNA binding increased, particularly after P-Tyr pre-treatment. These molecular effects of P-Tyr resulted in a reduction of residual DNA-damage after irradiation. Conclusions: Radioprotection by P-Tyr is mediated through its stimulation of nuclear EGFR transport and concurrent, but DNA-damage independent, activation of DNA-PK. Thus, subsequent irradiation results in increased binding of DNA-PK to DNA, improved DNA-repair and increased cell survival. c 2007 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 84 (2007) 328–334.
Keywords: Radiation; EGFR; DNA-PK; P-Tyr; DNA-repair
Phosphorylation of tyrosine residues of proteins plays a key role in signal transduction processes responsible for regulation of many essential cell reactions [28]. The crucial role of phosphorylation reactions is confirmed by the observation that phosphatases which reconstitute the dephosphorylated status are expressed in a high excess compared to the kinases responsible for the phosphorylation reaction [16]. Recently, it was reported that O-phospho-L-tyrosine (P-Tyr) inhibits the growth of breast and renal carcinoma cells in vitro after incubation for several days at millimolar concentrations [21]. An activation of phosphatases was suggested as molecular mechanism based on results indicating that the EGF-triggered tyrosine-phosphorylation of the epidermal growth factor receptor [21–23] is reduced following a P-Tyr pre-treatment. Usually, tyrosine-phosphorylation of these receptors is linked to signal transduction stimulating
the mitogenic activity of cells [27]. As a consequence inhibition of ligand induced phosphorylation, as reported by Mishra and Hamburger [22,23], was in agreement with the observed antiproliferative effect of P-Tyr. It was hypothesized that the effect of pTyr on EGFR phosphorylation could be due to the potential of P-Tyr to stimulate cellular phosphatases. However, it was never demonstrated that this interpretation is correct. We could show that P-Tyr treatment at micro-molar concentrations resulted in inhibition of PTPj and increased phosphorylation of EGFR. These molecular reactions were associated with a radioprotective effect, when cells were treated prior to irradiation [7]. Interestingly, radioprotection was only achieved in cells with a wild type TP53, which generates the possibility to protect surrounding normal tissue during radiotherapy, whereas tumor cells presenting
0167-8140/$ - see front matter c 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2007.07.006
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
TP53 mutations are not protected. This TP53 dependency was also observed for Bowman Birk protease inhibitor (BBI), a radioprotective peptide with 71 amino acids, characterized by an active centre with a phosphorylated Tyr residue [6]. In addition, we could show that for BBI-mediated radioprotection the activation of DNA-PK activity is essential for the effect. Consequently, DNA-repair quality was improved, measured as reduction of residual dicentric chromosomes [5]. These data suggest a clear effect of BBI and P-Tyr upon DNA-repair in general and presumably on DNA-double strand break repair especially. However, the molecular mode of action is unknown. Previous data reported by Mishra and Hamburger [21] and recent results from our laboratory indicate that P-Tyr interferes with EGFR phosphorylation and radiation-induced EGFR transport into the cell nucleus. This transport is involved in regulation of DNA-PK activity which is essential for DNA-repair [3,8]. Inhibition of EGFR transport by the EGFR specific, clinical relevant antibody Cetuximab [14], blocked transport and nuclear accumulation and inhibited radiation-induced activation of DNA-PK [4]. In the present study we investigated whether P-Tyr mediated phosphorylation of EGFR stimulates nuclear accumulation and consequently modifies repair of radiation-induced DNA-damage. We provide evidence that P-Tyr treatment prior to irradiation results in EGFR nuclear accumulation which is linked to DNA-PK phosphorylation. Thus, post irradiation P-Tyr stimulated DNA-PK activity is detected in complex with DNA which results in an improved DNA-repair.
Materials and methods Cell culture and irradiation Human bronchial carcinoma cells, designated A549 (ATCC), were used. Trypsinized cells were seeded for colony formation assay at a density of 300 per 9.6 cm2 well and 24 h after plating cells were treated with P-Tyr (10 lM) or/and Cetuximab (30 nM). Subsequently cells were irradiated with 225-kV photons (Gulmay RS 225) with a dose rate of 3 Gy/ min at 37 C. P-Tyr was purchased from Sigma and used at a concentration of 10 lM for 16 h before irradiation. EGFR-inhibitory antibody Cetuximab (Cetuximab) was supplied by Merck KG aA, Germany, and was used at a concentration of 30 nM either 1 h or 16 h before irradiation.
329
Subcellular fractionation Cytoplasmic and nuclear extracts were prepared according to the instructions of the NE-PER nuclear and cytoplasmic extraction kit (Pierce, Rockford, IL, USA).
Quantification of EGFR phosphorylation EGFR was immune-precipitated from 2 mg of cell lysate with monoclonal antibody clone 13 (BD Transduction Laboratories). The immune-complexes were washed and re-buffered with kinase buffer (10 mM Hepes, pH 7.5, 50 mM glycerophosphate, 50 mM NaCl, 10 mM MgCl2, 10 mM MnCl2, 5 lM ATP). Phosphorylation was quantified in the presence of 1 lCi c-[32P]ATP for 30 min at 30 C. After SDS–PAGE separation proteins were blotted, phosphorylated EGFR was visualized by fluorography. EGFR protein and phosphorylation at tyrosine residues were quantified with help of specific antibodies.
Quantification of RPTPj-phosphatase activity Cytoplasmic and nuclear extracts were prepared according to the manual of the NE-PER nuclear and cytoplasmic extraction kit (Pierce, Rockford). RPTPj was immune-precipitated with help of a specific antibody (Gene Tex) from cytoplasmic fraction. To quantify phosphatase activity a tyrosine-phosphorylated EGFR-peptide (Bachem AG) was added to a final concentration of 0.05 mM. Dephosphorylation reaction was performed in 50 ll assay buffer (50 mM TrisBase, 100 mM NaCl and 100 lg/ml bovine serum albumin) at RT for 15 min. Phosphatase reaction was stopped by addition of 100 ll BIOMOL-Green Reagent (Biomol) and free phosphate was quantified by measurement of absorbance at 620 nm.
Quantification of cH2AX-foci formation Cells were cultivated on CultureSlides (Becton–Dickinson), incubated with P-Tyr (10 lM) for 16 h, irradiated and fixed with 70% ice-cold ethanol after 24 h. For immune-fluorescence analysis cells were incubated with cH2AX antibody (Upstate, clone JBW301) (1:500) for 2 h at room temperature. Positive foci were visualized by incubation with a 1:500 dilution of Alexa488-labelled goat anti-mouse serum (Molecular Probes) for 30 min. Coverslips were mounted in Vectashield/DAPI (Vector Laboratories). For each data point from 300 to 500 nuclei were evaluated.
Western blot analysis and immune-precipitation Cell cultures were irradiated as described above, cells were lysed and proteins were resolved by SDS–PAGE. Western blotting was performed according to standard procedures. The primary antibodies were diluted as follows: anti-EGFR (BD Transduction Laboratories, clone 13) 1:1000; anti phospho-tyrosine (Santa Cruz, clone PY20) 1:1000; anti-DNA-PK (PharMingen, clone 4F10C5) 1:500; anti-DNA-PK phosphoThr No 2609 (Rockland) 1:1000. Quantification of binding was achieved by incubation with a secondary peroxidase-conjugated antibody with the ECL system (Amersham). EGFR was immune-precipitated from cytoplasmic and nuclear protein fractions obtained from 20 · 106 cells with antiEGFR antibody clone 13 (BD Transduction Laboratories).
Results Neutralization of the P-Tyr mediated radioprotection by incubation with the EGFR blocking antibody Cetuximab Ionizing radiation induced a dose dependent decline of surviving fraction in A549 cells. In accordance with earlier data [7] P-Tyr pre-incubation increased cell survival by about 50% (Fig. 1a). This radioprotective effect could completely be abolished by combined incubation with the EGFR blocking antibody Cetuximab (Fig. 1d) for 16 h. Cetuximab administration for only 1 h prior irradiation induced, as reported earlier [4], a marked radiosensitization (Fig. 1b).
330
P-Tyr stimulates DNA-repair via EGFR
Fig. 1. Neutralization of the P-Tyr associated radioprotective effect by incubation with the EGFR-inhibitory antibody Cetuximab. A549 were seeded at a density of 300 cells per 9.6 cm2 and after 24 h, cells were incubated with P-Tyr (10lM) for 16 h (a) or with Cetuximab (30 nM) for 1 h prior irradiation (b). During combined treatment cells were incubated with P-Tyr for 16 h and Cetuximab treatment was performed either for 16 h (c) or 1 h prior irradiation (d). Subsequently, cells were irradiated with 2, 4 and 6 Gy. After 10 days colonies were fixed and stained. Surviving fractions were calculated on the basis of colony counts and plating efficiency. Each value represents the mean of three independent experiments. Curves were fitted according to the linear quadratic model. Differences were considered as significant for p < 0.05 (Student’s ttest). Treatment with P-Tyr and P-Tyr combined with Cetuximab for 1 h was found as significant radioprotective, whereas treatment with Cetuximab alone for 1 h resulted in radiosensitization.
However, the same 1 h pre-irradiation Cetuximab treatment failed to abolish the radioprotective effect of P-Tyr, when cells were pre-treated for 16 h with P-Tyr. (Fig. 1c).
Stabilization and phosphorylation of EGFR induced by P-Tyr treatment To elucidate the role of EGFR during P-Tyr mediated radioprotection, we investigated phosphorylation status of the EGFR during incubation with P-Tyr. Already P-Tyr treatment for 3 h resulted in an increased amount of EGFR protein (Fig. 2a). The increased protein level persisted over the complete pre-incubation time of 16 h and was associated with elevated levels of EGFR protein phosphorylated at tyrosine residues at early time points (Fig. 2a). In addition, overall phosphorylation of EGFR after incubation with P-Tyr was visualized by metabolic labelling with [32P]ATP. Interestingly
overall phosphorylation of EGFR was stronger than phosphorylation at Tyr-residues and exhibited a slightly different kinetic. To elucidate the molecular mechanism of P-Tyr-induced EGFR phosphorylation, we quantified the effect of a P-Tyr incubation on RPTPj activity, known to dephosphorylate EGFR [29]. As shown in Fig. 2b, incubation of cells with P-Tyr for 16 h reduced RPTPj activity significantly.
P-Tyr mediated phosphorylation of nuclear EGFR at residue T654 and nuclear accumulation of EGFR To clarify the consequences of P-Tyr-induced EGFR phosphorylation and protein stabilization, we isolated cytoplasmic and nuclear proteins and analyzed EGFR specific phosphorylation patterns for differences associated with P-Tyr treatment. EGFR phosphorylations at well-characterized tyrosine residues Y1045 and Y1173 were modulated
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
331
treatment was associated with accumulation of nuclear EGFR protein (Fig. 3). To exclude contamination of nuclear fraction with cytoplasmic proteins, the absence of calnexin, an abundant cytoplasmic protein, was tested (Fig. 3).
P-Tyr pre-treatment increased EGFR- and DNA-PK protein in nucleus and strengthened radiationinduced phosphorylation of DNA-PK at T2609
Fig. 2. P-Tyr induced EGFR phosphorylation. (a) A549 cells were incubated with P-Tyr (10 lM) for various time intervals. Cells were harvested and EGFR immune-precipitation was performed. Precipitated proteins were incubated for 30 min at 30 C in presence of [32P]ATP and proteins were separated by SDS–PAGE and blotted. Phosphorylated proteins were visualized by fluorography. EGFR protein and EGFR phosphorylated at tyrosine residues were detected with help of a specific antibodies. As loading control actin in input was detected. The experiment was performed three times; shown are representative results. (b) A549 cells were incubated with P-Tyr (10 lM) for 16 h. Cells were harvested and RPTPj immune-precipitation was performed and RPTPj activity was quantified as triplicate. Asterisks indicate significant differences (Student’s t-test *p < 0.05).
only slightly following P-Tyr treatment (data not shown). However, phosphorylation at Thr residue No. 654 was increased significantly, when EGFR is localized within nucleus (Fig. 3). This increased phosphorylation in response to P-Tyr
Based on our reports that EGFR is transported into the nucleus upon irradiation [3] and that this process is involved in regulation of DNA-PK activity [4], we investigated these processes in response to P-Tyr-treatment or in combination with irradiation. Radiation induced a nuclear EGFR protein accumulation within 10 min after irradiation (Fig. 4a). P-Tyr pre-treatment for 16 h increased, as already shown (Fig. 3), the basal EGFR protein amount in the nucleus, however, additional radiation-induced nuclear transport was abolished (Fig. 4a) and EGFR protein persisted at a high level. Increased amounts of nuclear EGFR phosphorylated at T654 – as already observed after P-Tyr treatment alone (Fig. 3) – were detected with some delay compared to overall EGFR protein levels (Fig. 4a). This was true for radiation alone and even more pronounced for combined irradiation and P-Tyr-treatment. Moreover, detection of increased nuclear EGFR was associated with increased amounts of DNA-PK protein (Fig. 4a). Furthermore, DNA-PK phosphorylation at residue Thr2609, representing DNA-PK activity and known to be essential for DNA-repair processes [2,9], was clearly increased by P-Tyr pre-treatment (Fig. 4a). P-Tyr pre-treatment increased already basal Thr2609 phosphorylation of DNA-PK and the increase following irradiation was more pronounced (Fig. 4a). To answer the question, whether P-Tyr-activated DNA-PK is bound to DNA, we precipitated DNA from cells treated with P-Tyr for various time intervals (Fig. 4b). Nevertheless DNA-PK protein amount and phosphorylation at Thr2609 were increased upon treatment with P-Tyr alone (Fig. 4a, time point 0), DNA-PK binding to DNA was not increased (Fig. 4b). However, irradiation with 2 Gy resulted in a significant DNA binding of DNA-PK phosphorylated at Thr2609 (Fig. 4b). Pre-incubation with P-Tyr further increased binding of phosphorylated DNA-PK to DNA after irradiation.
P-Tyr pre-treatment improves DNA-repair after radiation exposure
Fig. 3. P-Tyr induced EGFR stabilization and phosphorylation at residue T654. Confluent A549 cells were incubated with P-Tyr (10 lM) for various times. Cytoplasmic and nuclear proteins were isolated from 20 · 106 cells, EGFR immune-precipitation was performed and proteins were separated by SDS–PAGE. After transfer to nitrocellulose, amounts of T654 phosphorylated EGFR and EGFR were quantified with help of specific antibodies. As loading controls nuclear protein lamin B1 and cytoplasmic protein calnexin were detected. Experiments were performed three times; shown are representative results.
To elucidate the effect of P-Tyr associated modulation of EGFR and DNA-PK phosphorylations upon DNA-repair processes, we detected residual DNA-damages 24 h after irradiation. Irradiation induced a dose dependent increase in residual cH2AX foci (Fig. 5). P-Tyr treatment had no significant impact upon basal amount of residual DNA-damage in the absence of irradiation, however, pre-incubation with P-Tyr reduced the amount of residual cH2AX foci per nucleus to about 50% at all doses applied (Fig. 5).
Discussion Already several years ago Mishra and Hamburger [21] reported that incubation of tumor cells with P-Tyr results
332
P-Tyr stimulates DNA-repair via EGFR
Fig. 4. P-Tyr mediated modulation of radiation-induced nuclear EGFR accumulation. (a) A549 cells were incubated with P-Tyr (10 lM) for 16 h and subsequently irradiated with 4 Gy. At the time points given after irradiation nuclear proteins were isolated, separated by SDS–PAGE and blotted. With help of specific antibodies the amounts of EGFR, EGFR phosphorylated at T654, DNA-PK and DNA-PK phosphorylated at T2609 were quantified. (b) P-Tyr induced phosphorylation of DNA-PK at residue T2609 and its binding to DNA after radiation. Cells were incubated with P-Tyr for the time intervals given, proteins were cross-linked to DNA by incubation with 1% formaldehyde and DNA/protein complexes were precipitated with isopropanol. Complexes were separated by SDS–PAGE and after Western blotting DNA-PK phosphorylated at T2609 was identified with help of specific antibody. Cells were pre-incubated with P-Tyr for 16 h and following irradiation DNA binding of T2609 phosphorylated DNA-PK was quantified as described above. The experiment was performed three times; shown are representative results.
in growth inhibition. This antiproliferative effect of P-Tyr is discussed to be due to inhibition of EGF-induced EGFR phosphorylation. EGFR phosphorylation usually results in generation of docking sites for several adapter proteins involved in EGFR-linked signalling [10]. To silence EGFR-signalling the receptor is internalized by means of clathrin coated pits and degraded or dephosphorylated and recycled to plasma membrane [12]. However, EGFR phosphorylation is also observed after irradiation or treatment with other cellular stressors, e.g. H2O2, [25]. This ligand independent phosphorylation is suggested to be the consequence of phosphatase inhibition [13]. EGFR phosphorylation results in caveolin 1 dependent internalization and sorts EGFR into a perinuclear compartment, where the EGFR remains stable in a functional state [12]. Yet, the functional consequence of this perinuclear receptor localization is still unclear and has to be elucidated. Meanwhile it is well accepted that EGFR is one of the key players during regulation of cellular stress responses [15,26]. It is proven that EGFR not only regulates cell proliferation, but is also important for the regulation of cell survival and DNA-repair [15,26,27]. Thus, blockage of EGFR
function by the EGFR-specific monoclonal antibody Cetuximab enhances radiation sensitivity by affecting cell survival most likely through inhibiting DNA-repair. Interestingly, P-Tyr induced EGFR-phosphorylation already after 3 h incubation. The effect of P-Tyr incubation upon phosphorylation of EGFR is correlated with a P-Tyr associated inhibition of RPTPj activity. This observation contradicts the hypothesis of Mishra and Hamburger [23] that P-Tyr acts as a general activator of PTPs. Several points can be discussed as explanation for this contradiction. Mishra and Hamburger used different cell systems, very long incubation times, very high concentrations and as end point EGF-induced EGFR Tyrphosphorylation. However, we favour the explanation that P-Tyr-induced phosphorylation of EGFR is associated with caveolin-linked receptor internalization, which blocks additional ligand dependent EGFR activation and subsequent receptor degradation [12]. Indeed we observed a P-Tyr induced protein stabilization of EGFR. Furthermore, the P-Tyr-induced EGFR internalization would explain the antiproliferative effect of P-Tyr observed in certain tumor cell lines [21]. P-Tyr incubation over several days – as performed by Mishra and Hamburger – would reduce the
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
Fig. 5. P-Tyr affects residual DNA-damage. A549 cells were treated with P-Tyr (10 lM) for 16 h. Subsequently cells were irradiated with 2, 4 and 6 Gy and after 24 h cells were fixed. Residual damage was visualized by incubation with antibody directed against cH2AX. Each bar represents the mean ± SE of residual repair foci positive for cH2AX per cell. For each data point from 300 to 500 nuclei were evaluated. Asterisks indicate significant differences (Student’s t-test *p < 0.05).
amount of cell membrane associated EGFR sensitive to EGFinduced phosphorylation. This would result in an overall reduction of EGFR phosphorylation over the time and can be misinterpreted as an activation of PTPs. Moreover, as demonstrated already earlier, we provide evidence that P-Tyr-induced phosphorylation of EGFR is associated with a pronounced radioprotective effect [7]. We found out that P-Tyr stimulated EGFR protein is in complex with activated caveolin 1 (data not shown), which suggests that the EGFR is internalized by means of caveolae and protected by that from degradation [12]. Most interestingly the P-Tyr mediated EGFR stabilization in the cytoplasm seems to be linked to nuclear transport of EGFR. A similar observation is reported by Lo et al. for EGF-triggered EGFR transport into the nucleus [18]. Nuclear transport of EGFR could also be observed after irradiation of cells and is linked to regulation of DNA-repair processes [5]. Although we failed to correlate nuclear EGFR amounts to phosphorylation at specific tyrosine residues (e.g. Y1045 and Y1173; data not shown) we postulate phosphorylation at other so far not identified tyrosine residues within the EGFR sequence. However, we were able to correlate the observed general phosphorylation of EGFR and nuclear localization after pTyr treatment with phosphorylation at residue Threonine No. 654. It is reported that phosphorylation at this site via PKC is important for internalization of membrane bound EGFR [20]. Most interestingly Lin et al. [17] identified a putative nuclear localization site (NLS) in the EGFR sequence which includes Thr654. Usually binding of karyopherins to the NLS regulates nuclear protein transport [24]. Karyopherin binding can be regulated by phosphorylation reactions [24]. Indeed we observed a radiation-associated phosphorylation of EGFR at Thr654 in the nucleus. In addition we showed previously [3] that radiation-induced transport of EGFR into the nucleus is associated with a complex formation between EGFR and karyopherin a. Interestingly,
333
P-Tyr-incubation induced per se EGFR phosphorylation at Thr654 in the nucleus and this event was correlated with accumulation of nuclear EGFR. We assume that phosphorylation of EGFR at T654 dissociates karyopherin a from EGFR. This would explain the nuclear accumulation of EGFR following P-Tyr treatment, since karyopherin a has a nuclear export sequence to ensure recycling back to cytoplasm. Thus, dissociation of EGFR from karyopherin a would lead to accumulation of EGFR protein within the nucleus. This process would happen after irradiation, pTyr-treatment or the combination of both and would explain the increased amounts of EGFR-protein within the nuclear fraction in both cases. As reported earlier, nuclear transport of EGFR was associated with import of EGFR-bound proteins, such as Ku70/ Ku80 [3]. Moreover, DNA-dependent protein kinase (DNAPK), which is essential for repair of DNA-double strand breaks, could be found in nuclear complex with EGFR [1,3,8,11]. As shown herein, p-Tyr treatment markedly increased DNA-PK protein amount in the nucleus. Thus, these results implicate that pTyr affects cellular DNA-damage repair capacity through stimulation of nuclear EGFR transport and accumulation. In accordance with this, we observed that the radioprotective effect of P-Tyr in vitro can be abolished by treatment with the EGFR blocking antibody Cetuximab, when the antibody is administered for longer than 1-h time intervals. However, a short time incubation with Cetuximab for 1 h fails to revert processes essential for radioprotection mediated by P-Tyr. This observation suggests that P-Tyr induces stable processes, e.g. stabilization of proteins, as reported for TP53 [7], which confers radioprotection to the cell. As already described earlier, Cetuximab immobilizes internalized EGFR in the cytoplasm and prevents EGFR nuclear entry [4]. Thus, the inhibition of P-Tyr mediated radioprotection by Cetuximab treatment suggests a crucial role of nuclear EGFR during this process. A functional connection between nuclear EGFR and nuclear DNA-PK activity after radiation treatment at the level of DNA-DSB-repair was reported by us recently [4]. Thus, the presented observation that P-Tyr pre-incubation significantly affects DNA-PK protein and activity in the nucleus perfectly fits our previous observation, that P-Tyr can increase clonogenic survival after irradiation [7]. DNA-PK activity is essential for non-homologous end-joining, a repair process which starts immediately after radiation exposure. It is discussed that this enzyme is involved in damage removal by regulating Artemis endonuclease activity [9,19]. As reported herein, P-Tyr treatment seems to activate this repair system independent of radiation exposure and radiation-induced DNA-damage. This assumption is substantiated by the fact that we could not observe any induction of cH2AX positive foci after P-Tyr pre-treatment alone. Nevertheless, our data indicate that P-Tyr activated DNA-PK in non-irradiated cells (Thr2609 phosphorylated) was not bound to DNA, whereas radiation exposure triggered fast binding of Thr2609 phosphorylated DNA-PK to DNA. Thus, we report here the first time that activation of DNA-PK can happen via DNA independent processes. In case of DNA-damage this activated DNA-PK binds to DNA in a second step.
334
P-Tyr stimulates DNA-repair via EGFR
In summary, we provide evidence that the radioprotective activity of P-Tyr is mediated through stimulation of nuclear EGFR accumulation and EGFR dependent stimulation of DNA-PK in a DNA-damage independent manner. Consequently, when radiation-induced DNA-damage occurs, DNA-PK pre-activated by P-Tyr can bind to DNA immediately and DNA-repair is improved resulting in enhanced cell survival after radiation exposure. Acknowledgement This work was supported by Deutsche Krebshilfe Grant No. 106401. * Corresponding author. Klaus Dittmann, Division of Radiobiology and Molecular Environmental Research, Department of Radiation Oncology, Eberhard-Karls-University, Ro ¨ntgenweg 11, 72076 Tu ¨bingen, Germany. E-mail address: klaus.dittmann@uni-tuebin gen.de Received 26 January 2007; received in revised form 22 June 2007; accepted 17 July 2007; Available online 21 August 2007
References [1] Bandyopadhyay D, Mandal M, Adam L, et al. Physical interaction between epidermal growth factor receptor and DNAdependent protein kinase in mammalian cells. J Biol Chem 1998;273:1568–73. [2] Chen BP, Chan DW, Kobayashi J, et al. Cell cycle dependence of DNA-dependent protein kinase phosphorylation in response to DNA double strand breaks. J Biol Chem 2005;280:14709–15. [3] Dittmann K, Mayer C, Fehrenbacher B, et al. Radiationinduced epidermal growth factor receptor nuclear import is linked to activation of DNA-dependent protein kinase. J Biol Chem 2005;280:31182–9. [4] Dittmann K, Mayer C, Rodemann HP. Inhibition of radiationinduced EGFR nuclear import by C225 (Cetuximab) suppresses DNA-PK activity. Radiother Oncol 2005;76:157–61. [5] Dittmann K, Virsik-Ko ¨pp P, Mayer C, et al. Bowman-Birk protease inhibitor activates DNA-dependent protein kinase and reduces formation of radiation-induced dicentric chromosomes. Int J Radiat Biol 2003;79:801–8. [6] Dittmann KH, Gueven N, Mayer C, et al. The presence of wildtype TP53 is necessary for the radioprotective effect of the Bowman-Birk proteinase inhibitor in normal fibroblasts. Radiat Res 1998;150:648–55. [7] Dittmann KH, Mayer C, Rodemann HP. O-phospho-L-tyrosine protects TP53 wild-type cells against ionizing radiation. Int J Cancer 2001;96:1–6. [8] Friedmann BJ, Caplin M, Savic B, et al. Interaction of the epidermal growth factor receptor and the DNA-dependent protein kinase pathway following gefitinib treatment. Mol Cancer Ther 2006;5:209–18. [9] Goodarzi AA, Yu Y, Riballo E, et al. DNA-PK autophosphorylation facilitates Artemis endonuclease activity. EMBO J 2006;25:3880–9. [10] Herbst RS. Review of epidermal growth factor receptor biology. Int J Radiat Oncol Biol Phys 2004;59:21–6. [11] Huang SM, Bock JM, Harari PM. Epidermal growth factor receptor blockade with C225 modulates proliferation, apoptosis, and radiosensitivity in squamous cell carcinomas of the head and neck. Cancer Res 1999;59:1935–40.
[12] Khan EM, Heidinger JM, Levy M, et al. Epidermal growth factor receptor exposed to oxidative stress undergoes Src- and caveolin-1-dependent perinuclear trafficking. J Biol Chem 2006;281:14486–93. [13] Knebel A, Rahmsdorf HJ, Ullrich A, Herrlich P. Dephosphorylation of receptor tyrosine kinases as target of regulation by radiation, oxidants or alkylating agents. EMBO J 1996;15:5314–25. [14] Krause M, Schutze C, Petersen C, et al. Different classes of EGFR inhibitors may have different potential to improve local tumour control after fractionated irradiation: a study on C225 in FaDu hSCC. Radiother Oncol 2005;74:109–15. [15] Lammering G. Molecular predictor and promising target: will EGFR now become a star in radiotherapy? Radiother Oncol 2005;74:89–91. [16] Lau KH, Baylink DJ. Phosphotyrosyl protein phosphatases: potential regulators of cell proliferation and differentiation. Crit Rev Oncog 1993;4:451–71. [17] Lin SY, Makino K, Xia W, et al. Nuclear localization of EGF receptor and its potential new role as a transcription factor. Nat Cell Biol 2001;3:802–8. [18] Lo HW, Ali-Seyed M, Wu Y, et al. Nuclear-cytoplasmic transport of EGFR involves receptor endocytosis, importin beta1 and CRM1. J Cell Biochem 2006;98:1570–83. [19] Lo ¨brich M, Jeggo PA. The two edges of the ATM sword: cooperation between repair and checkpoint functions. Radiother Oncol 2005;76:112–8. [20] Lund KA, Lazar CS, Chen WS, et al. Phosphorylation of the epidermal growth factor receptor at threonine 654 inhibits ligand-induced internalization and down-regulation. J Biol Chem 1990;265:20517–23. [21] Mishra S, Hamburger AW. Association of inhibition of cell growth by O-phospho-L-tyrosine with decreased tyrosine phosphorylation. Cancer Lett 1996;102:65–71. [22] Mishra S, Hamburger AW. Exogenous phosphotyrosine modulates epidermal growth factor receptor tyrosine phosphorylation. Carcinogenesis 1993;14:269–73. [23] Mishra S, Hamburger AW. O-phospho-L-tyrosine inhibits cellular growth by activating protein tyrosine phosphatases. Cancer Res 1993;53:557–63. [24] Mosammaparast N, Pemberton LF. Karyopherins: from nucleartransport mediators to nuclear-function regulators. Trends Cell Biol 2004;14:547–56. [25] Sturla LM, Amorino G, Alexander MS, et al. Requirement of Tyr-992 and Tyr-1173 in phosphorylation of the epidermal growth factor receptor by ionizing radiation and modulation by SHP2. J Biol Chem 2005;280:14597–604. [26] Toulany M, Dittmann K, Baumann M, Rodemann HP. Radiosensitization of Ras-mutated human tumor cells in vitro by the specific EGF receptor antagonist BIBX1382BS. Radiother Oncol 2005;74:117–29. [27] Toulany M, Dittmann K, Kru ¨ger M, et al. Radioresistance of KRas mutated human tumor cells is mediated through EGFRdependent activation of PI3K-AKT pathway. Radiother Oncol 2005;76:143–50. [28] Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell 1990;61:203–12. [29] Xu Y, Shao Y, Voorhees JJ, Fisher GJ. Oxidative inhibition of receptor-type protein-tyrosine phosphatase kappa by ultraviolet irradiation activates epidermal growth factor receptor in human keratinocytes. J Biol Chem 2006;281:27389–97.
Experimental radiobiology
The radioprotector O-phospho-tyrosine stimulates DNA-repair via epidermal growth factor receptor- and DNA-dependent kinase phosphorylation Klaus Dittmanna,*, Claus Mayera, Gabriele Wannera, Rainer Kehlbachb, H. Peter Rodemanna a
Division of Radiobiology and Molecular Environmental Research, Department of Radiation Oncology, and bDepartment of ¨bingen, Tu ¨bingen, Germany Radiology, University of Tu
Abstract Background and purpose: Purpose of the study was to elucidate the underlying molecular mechanism of the radioprotector O-phospho-tyrosine (P-Tyr). Methods: Molecular effects of P-Tyr at the level of EGFR responses were investigated in vitro with bronchial carcinoma cell line A549. Nuclear EGFR transport and DNA-PK activation were quantified after Western blotting. Residual DNAdamages were quantified by help of cH2AX focus assay. Results: As determined by dose–response curves, treatment of cells with P-Tyr for 16 h before irradiation results in radioprotection. Simultaneous treatment with EGFR blocking antibody Cetuximab abolished P-Tyr associated radioprotection. At the molecular level P-Tyr mediated a general phosphorylation of EGFR and a pronounced phosphorylation of nuclear EGFR at residue Thr No. 654, also observed after treatment with ionizing radiation. This phosphorylation was associated with nuclear EGFR accumulation. Moreover, P-Tyr-triggered EGFR nuclear accumulation was associated with phosphorylation of DNA-PK at Thr 2609. This activated form of DNA-PK was not DNA associated, but after radiation, DNA binding increased, particularly after P-Tyr pre-treatment. These molecular effects of P-Tyr resulted in a reduction of residual DNA-damage after irradiation. Conclusions: Radioprotection by P-Tyr is mediated through its stimulation of nuclear EGFR transport and concurrent, but DNA-damage independent, activation of DNA-PK. Thus, subsequent irradiation results in increased binding of DNA-PK to DNA, improved DNA-repair and increased cell survival. c 2007 Elsevier Ireland Ltd. All rights reserved. Radiotherapy and Oncology 84 (2007) 328–334.
Keywords: Radiation; EGFR; DNA-PK; P-Tyr; DNA-repair
Phosphorylation of tyrosine residues of proteins plays a key role in signal transduction processes responsible for regulation of many essential cell reactions [28]. The crucial role of phosphorylation reactions is confirmed by the observation that phosphatases which reconstitute the dephosphorylated status are expressed in a high excess compared to the kinases responsible for the phosphorylation reaction [16]. Recently, it was reported that O-phospho-L-tyrosine (P-Tyr) inhibits the growth of breast and renal carcinoma cells in vitro after incubation for several days at millimolar concentrations [21]. An activation of phosphatases was suggested as molecular mechanism based on results indicating that the EGF-triggered tyrosine-phosphorylation of the epidermal growth factor receptor [21–23] is reduced following a P-Tyr pre-treatment. Usually, tyrosine-phosphorylation of these receptors is linked to signal transduction stimulating
the mitogenic activity of cells [27]. As a consequence inhibition of ligand induced phosphorylation, as reported by Mishra and Hamburger [22,23], was in agreement with the observed antiproliferative effect of P-Tyr. It was hypothesized that the effect of pTyr on EGFR phosphorylation could be due to the potential of P-Tyr to stimulate cellular phosphatases. However, it was never demonstrated that this interpretation is correct. We could show that P-Tyr treatment at micro-molar concentrations resulted in inhibition of PTPj and increased phosphorylation of EGFR. These molecular reactions were associated with a radioprotective effect, when cells were treated prior to irradiation [7]. Interestingly, radioprotection was only achieved in cells with a wild type TP53, which generates the possibility to protect surrounding normal tissue during radiotherapy, whereas tumor cells presenting
0167-8140/$ - see front matter c 2007 Elsevier Ireland Ltd. All rights reserved. doi:10.1016/j.radonc.2007.07.006
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
TP53 mutations are not protected. This TP53 dependency was also observed for Bowman Birk protease inhibitor (BBI), a radioprotective peptide with 71 amino acids, characterized by an active centre with a phosphorylated Tyr residue [6]. In addition, we could show that for BBI-mediated radioprotection the activation of DNA-PK activity is essential for the effect. Consequently, DNA-repair quality was improved, measured as reduction of residual dicentric chromosomes [5]. These data suggest a clear effect of BBI and P-Tyr upon DNA-repair in general and presumably on DNA-double strand break repair especially. However, the molecular mode of action is unknown. Previous data reported by Mishra and Hamburger [21] and recent results from our laboratory indicate that P-Tyr interferes with EGFR phosphorylation and radiation-induced EGFR transport into the cell nucleus. This transport is involved in regulation of DNA-PK activity which is essential for DNA-repair [3,8]. Inhibition of EGFR transport by the EGFR specific, clinical relevant antibody Cetuximab [14], blocked transport and nuclear accumulation and inhibited radiation-induced activation of DNA-PK [4]. In the present study we investigated whether P-Tyr mediated phosphorylation of EGFR stimulates nuclear accumulation and consequently modifies repair of radiation-induced DNA-damage. We provide evidence that P-Tyr treatment prior to irradiation results in EGFR nuclear accumulation which is linked to DNA-PK phosphorylation. Thus, post irradiation P-Tyr stimulated DNA-PK activity is detected in complex with DNA which results in an improved DNA-repair.
Materials and methods Cell culture and irradiation Human bronchial carcinoma cells, designated A549 (ATCC), were used. Trypsinized cells were seeded for colony formation assay at a density of 300 per 9.6 cm2 well and 24 h after plating cells were treated with P-Tyr (10 lM) or/and Cetuximab (30 nM). Subsequently cells were irradiated with 225-kV photons (Gulmay RS 225) with a dose rate of 3 Gy/ min at 37 C. P-Tyr was purchased from Sigma and used at a concentration of 10 lM for 16 h before irradiation. EGFR-inhibitory antibody Cetuximab (Cetuximab) was supplied by Merck KG aA, Germany, and was used at a concentration of 30 nM either 1 h or 16 h before irradiation.
329
Subcellular fractionation Cytoplasmic and nuclear extracts were prepared according to the instructions of the NE-PER nuclear and cytoplasmic extraction kit (Pierce, Rockford, IL, USA).
Quantification of EGFR phosphorylation EGFR was immune-precipitated from 2 mg of cell lysate with monoclonal antibody clone 13 (BD Transduction Laboratories). The immune-complexes were washed and re-buffered with kinase buffer (10 mM Hepes, pH 7.5, 50 mM glycerophosphate, 50 mM NaCl, 10 mM MgCl2, 10 mM MnCl2, 5 lM ATP). Phosphorylation was quantified in the presence of 1 lCi c-[32P]ATP for 30 min at 30 C. After SDS–PAGE separation proteins were blotted, phosphorylated EGFR was visualized by fluorography. EGFR protein and phosphorylation at tyrosine residues were quantified with help of specific antibodies.
Quantification of RPTPj-phosphatase activity Cytoplasmic and nuclear extracts were prepared according to the manual of the NE-PER nuclear and cytoplasmic extraction kit (Pierce, Rockford). RPTPj was immune-precipitated with help of a specific antibody (Gene Tex) from cytoplasmic fraction. To quantify phosphatase activity a tyrosine-phosphorylated EGFR-peptide (Bachem AG) was added to a final concentration of 0.05 mM. Dephosphorylation reaction was performed in 50 ll assay buffer (50 mM TrisBase, 100 mM NaCl and 100 lg/ml bovine serum albumin) at RT for 15 min. Phosphatase reaction was stopped by addition of 100 ll BIOMOL-Green Reagent (Biomol) and free phosphate was quantified by measurement of absorbance at 620 nm.
Quantification of cH2AX-foci formation Cells were cultivated on CultureSlides (Becton–Dickinson), incubated with P-Tyr (10 lM) for 16 h, irradiated and fixed with 70% ice-cold ethanol after 24 h. For immune-fluorescence analysis cells were incubated with cH2AX antibody (Upstate, clone JBW301) (1:500) for 2 h at room temperature. Positive foci were visualized by incubation with a 1:500 dilution of Alexa488-labelled goat anti-mouse serum (Molecular Probes) for 30 min. Coverslips were mounted in Vectashield/DAPI (Vector Laboratories). For each data point from 300 to 500 nuclei were evaluated.
Western blot analysis and immune-precipitation Cell cultures were irradiated as described above, cells were lysed and proteins were resolved by SDS–PAGE. Western blotting was performed according to standard procedures. The primary antibodies were diluted as follows: anti-EGFR (BD Transduction Laboratories, clone 13) 1:1000; anti phospho-tyrosine (Santa Cruz, clone PY20) 1:1000; anti-DNA-PK (PharMingen, clone 4F10C5) 1:500; anti-DNA-PK phosphoThr No 2609 (Rockland) 1:1000. Quantification of binding was achieved by incubation with a secondary peroxidase-conjugated antibody with the ECL system (Amersham). EGFR was immune-precipitated from cytoplasmic and nuclear protein fractions obtained from 20 · 106 cells with antiEGFR antibody clone 13 (BD Transduction Laboratories).
Results Neutralization of the P-Tyr mediated radioprotection by incubation with the EGFR blocking antibody Cetuximab Ionizing radiation induced a dose dependent decline of surviving fraction in A549 cells. In accordance with earlier data [7] P-Tyr pre-incubation increased cell survival by about 50% (Fig. 1a). This radioprotective effect could completely be abolished by combined incubation with the EGFR blocking antibody Cetuximab (Fig. 1d) for 16 h. Cetuximab administration for only 1 h prior irradiation induced, as reported earlier [4], a marked radiosensitization (Fig. 1b).
330
P-Tyr stimulates DNA-repair via EGFR
Fig. 1. Neutralization of the P-Tyr associated radioprotective effect by incubation with the EGFR-inhibitory antibody Cetuximab. A549 were seeded at a density of 300 cells per 9.6 cm2 and after 24 h, cells were incubated with P-Tyr (10lM) for 16 h (a) or with Cetuximab (30 nM) for 1 h prior irradiation (b). During combined treatment cells were incubated with P-Tyr for 16 h and Cetuximab treatment was performed either for 16 h (c) or 1 h prior irradiation (d). Subsequently, cells were irradiated with 2, 4 and 6 Gy. After 10 days colonies were fixed and stained. Surviving fractions were calculated on the basis of colony counts and plating efficiency. Each value represents the mean of three independent experiments. Curves were fitted according to the linear quadratic model. Differences were considered as significant for p < 0.05 (Student’s ttest). Treatment with P-Tyr and P-Tyr combined with Cetuximab for 1 h was found as significant radioprotective, whereas treatment with Cetuximab alone for 1 h resulted in radiosensitization.
However, the same 1 h pre-irradiation Cetuximab treatment failed to abolish the radioprotective effect of P-Tyr, when cells were pre-treated for 16 h with P-Tyr. (Fig. 1c).
Stabilization and phosphorylation of EGFR induced by P-Tyr treatment To elucidate the role of EGFR during P-Tyr mediated radioprotection, we investigated phosphorylation status of the EGFR during incubation with P-Tyr. Already P-Tyr treatment for 3 h resulted in an increased amount of EGFR protein (Fig. 2a). The increased protein level persisted over the complete pre-incubation time of 16 h and was associated with elevated levels of EGFR protein phosphorylated at tyrosine residues at early time points (Fig. 2a). In addition, overall phosphorylation of EGFR after incubation with P-Tyr was visualized by metabolic labelling with [32P]ATP. Interestingly
overall phosphorylation of EGFR was stronger than phosphorylation at Tyr-residues and exhibited a slightly different kinetic. To elucidate the molecular mechanism of P-Tyr-induced EGFR phosphorylation, we quantified the effect of a P-Tyr incubation on RPTPj activity, known to dephosphorylate EGFR [29]. As shown in Fig. 2b, incubation of cells with P-Tyr for 16 h reduced RPTPj activity significantly.
P-Tyr mediated phosphorylation of nuclear EGFR at residue T654 and nuclear accumulation of EGFR To clarify the consequences of P-Tyr-induced EGFR phosphorylation and protein stabilization, we isolated cytoplasmic and nuclear proteins and analyzed EGFR specific phosphorylation patterns for differences associated with P-Tyr treatment. EGFR phosphorylations at well-characterized tyrosine residues Y1045 and Y1173 were modulated
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
331
treatment was associated with accumulation of nuclear EGFR protein (Fig. 3). To exclude contamination of nuclear fraction with cytoplasmic proteins, the absence of calnexin, an abundant cytoplasmic protein, was tested (Fig. 3).
P-Tyr pre-treatment increased EGFR- and DNA-PK protein in nucleus and strengthened radiationinduced phosphorylation of DNA-PK at T2609
Fig. 2. P-Tyr induced EGFR phosphorylation. (a) A549 cells were incubated with P-Tyr (10 lM) for various time intervals. Cells were harvested and EGFR immune-precipitation was performed. Precipitated proteins were incubated for 30 min at 30 C in presence of [32P]ATP and proteins were separated by SDS–PAGE and blotted. Phosphorylated proteins were visualized by fluorography. EGFR protein and EGFR phosphorylated at tyrosine residues were detected with help of a specific antibodies. As loading control actin in input was detected. The experiment was performed three times; shown are representative results. (b) A549 cells were incubated with P-Tyr (10 lM) for 16 h. Cells were harvested and RPTPj immune-precipitation was performed and RPTPj activity was quantified as triplicate. Asterisks indicate significant differences (Student’s t-test *p < 0.05).
only slightly following P-Tyr treatment (data not shown). However, phosphorylation at Thr residue No. 654 was increased significantly, when EGFR is localized within nucleus (Fig. 3). This increased phosphorylation in response to P-Tyr
Based on our reports that EGFR is transported into the nucleus upon irradiation [3] and that this process is involved in regulation of DNA-PK activity [4], we investigated these processes in response to P-Tyr-treatment or in combination with irradiation. Radiation induced a nuclear EGFR protein accumulation within 10 min after irradiation (Fig. 4a). P-Tyr pre-treatment for 16 h increased, as already shown (Fig. 3), the basal EGFR protein amount in the nucleus, however, additional radiation-induced nuclear transport was abolished (Fig. 4a) and EGFR protein persisted at a high level. Increased amounts of nuclear EGFR phosphorylated at T654 – as already observed after P-Tyr treatment alone (Fig. 3) – were detected with some delay compared to overall EGFR protein levels (Fig. 4a). This was true for radiation alone and even more pronounced for combined irradiation and P-Tyr-treatment. Moreover, detection of increased nuclear EGFR was associated with increased amounts of DNA-PK protein (Fig. 4a). Furthermore, DNA-PK phosphorylation at residue Thr2609, representing DNA-PK activity and known to be essential for DNA-repair processes [2,9], was clearly increased by P-Tyr pre-treatment (Fig. 4a). P-Tyr pre-treatment increased already basal Thr2609 phosphorylation of DNA-PK and the increase following irradiation was more pronounced (Fig. 4a). To answer the question, whether P-Tyr-activated DNA-PK is bound to DNA, we precipitated DNA from cells treated with P-Tyr for various time intervals (Fig. 4b). Nevertheless DNA-PK protein amount and phosphorylation at Thr2609 were increased upon treatment with P-Tyr alone (Fig. 4a, time point 0), DNA-PK binding to DNA was not increased (Fig. 4b). However, irradiation with 2 Gy resulted in a significant DNA binding of DNA-PK phosphorylated at Thr2609 (Fig. 4b). Pre-incubation with P-Tyr further increased binding of phosphorylated DNA-PK to DNA after irradiation.
P-Tyr pre-treatment improves DNA-repair after radiation exposure
Fig. 3. P-Tyr induced EGFR stabilization and phosphorylation at residue T654. Confluent A549 cells were incubated with P-Tyr (10 lM) for various times. Cytoplasmic and nuclear proteins were isolated from 20 · 106 cells, EGFR immune-precipitation was performed and proteins were separated by SDS–PAGE. After transfer to nitrocellulose, amounts of T654 phosphorylated EGFR and EGFR were quantified with help of specific antibodies. As loading controls nuclear protein lamin B1 and cytoplasmic protein calnexin were detected. Experiments were performed three times; shown are representative results.
To elucidate the effect of P-Tyr associated modulation of EGFR and DNA-PK phosphorylations upon DNA-repair processes, we detected residual DNA-damages 24 h after irradiation. Irradiation induced a dose dependent increase in residual cH2AX foci (Fig. 5). P-Tyr treatment had no significant impact upon basal amount of residual DNA-damage in the absence of irradiation, however, pre-incubation with P-Tyr reduced the amount of residual cH2AX foci per nucleus to about 50% at all doses applied (Fig. 5).
Discussion Already several years ago Mishra and Hamburger [21] reported that incubation of tumor cells with P-Tyr results
332
P-Tyr stimulates DNA-repair via EGFR
Fig. 4. P-Tyr mediated modulation of radiation-induced nuclear EGFR accumulation. (a) A549 cells were incubated with P-Tyr (10 lM) for 16 h and subsequently irradiated with 4 Gy. At the time points given after irradiation nuclear proteins were isolated, separated by SDS–PAGE and blotted. With help of specific antibodies the amounts of EGFR, EGFR phosphorylated at T654, DNA-PK and DNA-PK phosphorylated at T2609 were quantified. (b) P-Tyr induced phosphorylation of DNA-PK at residue T2609 and its binding to DNA after radiation. Cells were incubated with P-Tyr for the time intervals given, proteins were cross-linked to DNA by incubation with 1% formaldehyde and DNA/protein complexes were precipitated with isopropanol. Complexes were separated by SDS–PAGE and after Western blotting DNA-PK phosphorylated at T2609 was identified with help of specific antibody. Cells were pre-incubated with P-Tyr for 16 h and following irradiation DNA binding of T2609 phosphorylated DNA-PK was quantified as described above. The experiment was performed three times; shown are representative results.
in growth inhibition. This antiproliferative effect of P-Tyr is discussed to be due to inhibition of EGF-induced EGFR phosphorylation. EGFR phosphorylation usually results in generation of docking sites for several adapter proteins involved in EGFR-linked signalling [10]. To silence EGFR-signalling the receptor is internalized by means of clathrin coated pits and degraded or dephosphorylated and recycled to plasma membrane [12]. However, EGFR phosphorylation is also observed after irradiation or treatment with other cellular stressors, e.g. H2O2, [25]. This ligand independent phosphorylation is suggested to be the consequence of phosphatase inhibition [13]. EGFR phosphorylation results in caveolin 1 dependent internalization and sorts EGFR into a perinuclear compartment, where the EGFR remains stable in a functional state [12]. Yet, the functional consequence of this perinuclear receptor localization is still unclear and has to be elucidated. Meanwhile it is well accepted that EGFR is one of the key players during regulation of cellular stress responses [15,26]. It is proven that EGFR not only regulates cell proliferation, but is also important for the regulation of cell survival and DNA-repair [15,26,27]. Thus, blockage of EGFR
function by the EGFR-specific monoclonal antibody Cetuximab enhances radiation sensitivity by affecting cell survival most likely through inhibiting DNA-repair. Interestingly, P-Tyr induced EGFR-phosphorylation already after 3 h incubation. The effect of P-Tyr incubation upon phosphorylation of EGFR is correlated with a P-Tyr associated inhibition of RPTPj activity. This observation contradicts the hypothesis of Mishra and Hamburger [23] that P-Tyr acts as a general activator of PTPs. Several points can be discussed as explanation for this contradiction. Mishra and Hamburger used different cell systems, very long incubation times, very high concentrations and as end point EGF-induced EGFR Tyrphosphorylation. However, we favour the explanation that P-Tyr-induced phosphorylation of EGFR is associated with caveolin-linked receptor internalization, which blocks additional ligand dependent EGFR activation and subsequent receptor degradation [12]. Indeed we observed a P-Tyr induced protein stabilization of EGFR. Furthermore, the P-Tyr-induced EGFR internalization would explain the antiproliferative effect of P-Tyr observed in certain tumor cell lines [21]. P-Tyr incubation over several days – as performed by Mishra and Hamburger – would reduce the
K. Dittmann et al. / Radiotherapy and Oncology 84 (2007) 328–334
Fig. 5. P-Tyr affects residual DNA-damage. A549 cells were treated with P-Tyr (10 lM) for 16 h. Subsequently cells were irradiated with 2, 4 and 6 Gy and after 24 h cells were fixed. Residual damage was visualized by incubation with antibody directed against cH2AX. Each bar represents the mean ± SE of residual repair foci positive for cH2AX per cell. For each data point from 300 to 500 nuclei were evaluated. Asterisks indicate significant differences (Student’s t-test *p < 0.05).
amount of cell membrane associated EGFR sensitive to EGFinduced phosphorylation. This would result in an overall reduction of EGFR phosphorylation over the time and can be misinterpreted as an activation of PTPs. Moreover, as demonstrated already earlier, we provide evidence that P-Tyr-induced phosphorylation of EGFR is associated with a pronounced radioprotective effect [7]. We found out that P-Tyr stimulated EGFR protein is in complex with activated caveolin 1 (data not shown), which suggests that the EGFR is internalized by means of caveolae and protected by that from degradation [12]. Most interestingly the P-Tyr mediated EGFR stabilization in the cytoplasm seems to be linked to nuclear transport of EGFR. A similar observation is reported by Lo et al. for EGF-triggered EGFR transport into the nucleus [18]. Nuclear transport of EGFR could also be observed after irradiation of cells and is linked to regulation of DNA-repair processes [5]. Although we failed to correlate nuclear EGFR amounts to phosphorylation at specific tyrosine residues (e.g. Y1045 and Y1173; data not shown) we postulate phosphorylation at other so far not identified tyrosine residues within the EGFR sequence. However, we were able to correlate the observed general phosphorylation of EGFR and nuclear localization after pTyr treatment with phosphorylation at residue Threonine No. 654. It is reported that phosphorylation at this site via PKC is important for internalization of membrane bound EGFR [20]. Most interestingly Lin et al. [17] identified a putative nuclear localization site (NLS) in the EGFR sequence which includes Thr654. Usually binding of karyopherins to the NLS regulates nuclear protein transport [24]. Karyopherin binding can be regulated by phosphorylation reactions [24]. Indeed we observed a radiation-associated phosphorylation of EGFR at Thr654 in the nucleus. In addition we showed previously [3] that radiation-induced transport of EGFR into the nucleus is associated with a complex formation between EGFR and karyopherin a. Interestingly,
333
P-Tyr-incubation induced per se EGFR phosphorylation at Thr654 in the nucleus and this event was correlated with accumulation of nuclear EGFR. We assume that phosphorylation of EGFR at T654 dissociates karyopherin a from EGFR. This would explain the nuclear accumulation of EGFR following P-Tyr treatment, since karyopherin a has a nuclear export sequence to ensure recycling back to cytoplasm. Thus, dissociation of EGFR from karyopherin a would lead to accumulation of EGFR protein within the nucleus. This process would happen after irradiation, pTyr-treatment or the combination of both and would explain the increased amounts of EGFR-protein within the nuclear fraction in both cases. As reported earlier, nuclear transport of EGFR was associated with import of EGFR-bound proteins, such as Ku70/ Ku80 [3]. Moreover, DNA-dependent protein kinase (DNAPK), which is essential for repair of DNA-double strand breaks, could be found in nuclear complex with EGFR [1,3,8,11]. As shown herein, p-Tyr treatment markedly increased DNA-PK protein amount in the nucleus. Thus, these results implicate that pTyr affects cellular DNA-damage repair capacity through stimulation of nuclear EGFR transport and accumulation. In accordance with this, we observed that the radioprotective effect of P-Tyr in vitro can be abolished by treatment with the EGFR blocking antibody Cetuximab, when the antibody is administered for longer than 1-h time intervals. However, a short time incubation with Cetuximab for 1 h fails to revert processes essential for radioprotection mediated by P-Tyr. This observation suggests that P-Tyr induces stable processes, e.g. stabilization of proteins, as reported for TP53 [7], which confers radioprotection to the cell. As already described earlier, Cetuximab immobilizes internalized EGFR in the cytoplasm and prevents EGFR nuclear entry [4]. Thus, the inhibition of P-Tyr mediated radioprotection by Cetuximab treatment suggests a crucial role of nuclear EGFR during this process. A functional connection between nuclear EGFR and nuclear DNA-PK activity after radiation treatment at the level of DNA-DSB-repair was reported by us recently [4]. Thus, the presented observation that P-Tyr pre-incubation significantly affects DNA-PK protein and activity in the nucleus perfectly fits our previous observation, that P-Tyr can increase clonogenic survival after irradiation [7]. DNA-PK activity is essential for non-homologous end-joining, a repair process which starts immediately after radiation exposure. It is discussed that this enzyme is involved in damage removal by regulating Artemis endonuclease activity [9,19]. As reported herein, P-Tyr treatment seems to activate this repair system independent of radiation exposure and radiation-induced DNA-damage. This assumption is substantiated by the fact that we could not observe any induction of cH2AX positive foci after P-Tyr pre-treatment alone. Nevertheless, our data indicate that P-Tyr activated DNA-PK in non-irradiated cells (Thr2609 phosphorylated) was not bound to DNA, whereas radiation exposure triggered fast binding of Thr2609 phosphorylated DNA-PK to DNA. Thus, we report here the first time that activation of DNA-PK can happen via DNA independent processes. In case of DNA-damage this activated DNA-PK binds to DNA in a second step.
334
P-Tyr stimulates DNA-repair via EGFR
In summary, we provide evidence that the radioprotective activity of P-Tyr is mediated through stimulation of nuclear EGFR accumulation and EGFR dependent stimulation of DNA-PK in a DNA-damage independent manner. Consequently, when radiation-induced DNA-damage occurs, DNA-PK pre-activated by P-Tyr can bind to DNA immediately and DNA-repair is improved resulting in enhanced cell survival after radiation exposure. Acknowledgement This work was supported by Deutsche Krebshilfe Grant No. 106401. * Corresponding author. Klaus Dittmann, Division of Radiobiology and Molecular Environmental Research, Department of Radiation Oncology, Eberhard-Karls-University, Ro ¨ntgenweg 11, 72076 Tu ¨bingen, Germany. E-mail address: klaus.dittmann@uni-tuebin gen.de Received 26 January 2007; received in revised form 22 June 2007; accepted 17 July 2007; Available online 21 August 2007
References [1] Bandyopadhyay D, Mandal M, Adam L, et al. Physical interaction between epidermal growth factor receptor and DNAdependent protein kinase in mammalian cells. J Biol Chem 1998;273:1568–73. [2] Chen BP, Chan DW, Kobayashi J, et al. Cell cycle dependence of DNA-dependent protein kinase phosphorylation in response to DNA double strand breaks. J Biol Chem 2005;280:14709–15. [3] Dittmann K, Mayer C, Fehrenbacher B, et al. Radiationinduced epidermal growth factor receptor nuclear import is linked to activation of DNA-dependent protein kinase. J Biol Chem 2005;280:31182–9. [4] Dittmann K, Mayer C, Rodemann HP. Inhibition of radiationinduced EGFR nuclear import by C225 (Cetuximab) suppresses DNA-PK activity. Radiother Oncol 2005;76:157–61. [5] Dittmann K, Virsik-Ko ¨pp P, Mayer C, et al. Bowman-Birk protease inhibitor activates DNA-dependent protein kinase and reduces formation of radiation-induced dicentric chromosomes. Int J Radiat Biol 2003;79:801–8. [6] Dittmann KH, Gueven N, Mayer C, et al. The presence of wildtype TP53 is necessary for the radioprotective effect of the Bowman-Birk proteinase inhibitor in normal fibroblasts. Radiat Res 1998;150:648–55. [7] Dittmann KH, Mayer C, Rodemann HP. O-phospho-L-tyrosine protects TP53 wild-type cells against ionizing radiation. Int J Cancer 2001;96:1–6. [8] Friedmann BJ, Caplin M, Savic B, et al. Interaction of the epidermal growth factor receptor and the DNA-dependent protein kinase pathway following gefitinib treatment. Mol Cancer Ther 2006;5:209–18. [9] Goodarzi AA, Yu Y, Riballo E, et al. DNA-PK autophosphorylation facilitates Artemis endonuclease activity. EMBO J 2006;25:3880–9. [10] Herbst RS. Review of epidermal growth factor receptor biology. Int J Radiat Oncol Biol Phys 2004;59:21–6. [11] Huang SM, Bock JM, Harari PM. Epidermal growth factor receptor blockade with C225 modulates proliferation, apoptosis, and radiosensitivity in squamous cell carcinomas of the head and neck. Cancer Res 1999;59:1935–40.
[12] Khan EM, Heidinger JM, Levy M, et al. Epidermal growth factor receptor exposed to oxidative stress undergoes Src- and caveolin-1-dependent perinuclear trafficking. J Biol Chem 2006;281:14486–93. [13] Knebel A, Rahmsdorf HJ, Ullrich A, Herrlich P. Dephosphorylation of receptor tyrosine kinases as target of regulation by radiation, oxidants or alkylating agents. EMBO J 1996;15:5314–25. [14] Krause M, Schutze C, Petersen C, et al. Different classes of EGFR inhibitors may have different potential to improve local tumour control after fractionated irradiation: a study on C225 in FaDu hSCC. Radiother Oncol 2005;74:109–15. [15] Lammering G. Molecular predictor and promising target: will EGFR now become a star in radiotherapy? Radiother Oncol 2005;74:89–91. [16] Lau KH, Baylink DJ. Phosphotyrosyl protein phosphatases: potential regulators of cell proliferation and differentiation. Crit Rev Oncog 1993;4:451–71. [17] Lin SY, Makino K, Xia W, et al. Nuclear localization of EGF receptor and its potential new role as a transcription factor. Nat Cell Biol 2001;3:802–8. [18] Lo HW, Ali-Seyed M, Wu Y, et al. Nuclear-cytoplasmic transport of EGFR involves receptor endocytosis, importin beta1 and CRM1. J Cell Biochem 2006;98:1570–83. [19] Lo ¨brich M, Jeggo PA. The two edges of the ATM sword: cooperation between repair and checkpoint functions. Radiother Oncol 2005;76:112–8. [20] Lund KA, Lazar CS, Chen WS, et al. Phosphorylation of the epidermal growth factor receptor at threonine 654 inhibits ligand-induced internalization and down-regulation. J Biol Chem 1990;265:20517–23. [21] Mishra S, Hamburger AW. Association of inhibition of cell growth by O-phospho-L-tyrosine with decreased tyrosine phosphorylation. Cancer Lett 1996;102:65–71. [22] Mishra S, Hamburger AW. Exogenous phosphotyrosine modulates epidermal growth factor receptor tyrosine phosphorylation. Carcinogenesis 1993;14:269–73. [23] Mishra S, Hamburger AW. O-phospho-L-tyrosine inhibits cellular growth by activating protein tyrosine phosphatases. Cancer Res 1993;53:557–63. [24] Mosammaparast N, Pemberton LF. Karyopherins: from nucleartransport mediators to nuclear-function regulators. Trends Cell Biol 2004;14:547–56. [25] Sturla LM, Amorino G, Alexander MS, et al. Requirement of Tyr-992 and Tyr-1173 in phosphorylation of the epidermal growth factor receptor by ionizing radiation and modulation by SHP2. J Biol Chem 2005;280:14597–604. [26] Toulany M, Dittmann K, Baumann M, Rodemann HP. Radiosensitization of Ras-mutated human tumor cells in vitro by the specific EGF receptor antagonist BIBX1382BS. Radiother Oncol 2005;74:117–29. [27] Toulany M, Dittmann K, Kru ¨ger M, et al. Radioresistance of KRas mutated human tumor cells is mediated through EGFRdependent activation of PI3K-AKT pathway. Radiother Oncol 2005;76:143–50. [28] Ullrich A, Schlessinger J. Signal transduction by receptors with tyrosine kinase activity. Cell 1990;61:203–12. [29] Xu Y, Shao Y, Voorhees JJ, Fisher GJ. Oxidative inhibition of receptor-type protein-tyrosine phosphatase kappa by ultraviolet irradiation activates epidermal growth factor receptor in human keratinocytes. J Biol Chem 2006;281:27389–97.