Inhibiting Tyrosine Phosphorylation of Protein Kinase Cδ (PKCδ) Protects the Salivary Gland from Radiation Damage*

Background: Nuclear import of protein kinase Cδ is required for DNA-damage-induced apoptosis. Results: c-Src and c-Abl phosphorylate PKCδ to regulate nuclear import. Tyrosine kinase inhibitors block nuclear translocation of PKCδ and suppress apoptosis. Conclusion: Tyrosine kinase inhibitors can regulate the pro-apoptotic function of protein kinase Cδ. Significance: Tyrosine kinase inhibitors may improve the quality of life in cancer patients receiving radiation therapy. Radiation therapy for head and neck cancer can result in extensive damage to normal adjacent tissues such as the salivary gland and oral mucosa. We have shown previously that tyrosine phosphorylation at Tyr-64 and Tyr-155 activates PKCδ in response to apoptotic stimuli by facilitating its nuclear import. Here we have identified the tyrosine kinases that mediate activation of PKCδ in apoptotic cells and have explored the use of tyrosine kinase inhibitors for suppression of irradiation-induced apoptosis. We identify the damage-inducible kinase, c-Abl, as the PKCδ Tyr-155 kinase and c-Src as the Tyr-64 kinase. Depletion of c-Abl or c-Src with shRNA decreased irradiation- and etoposide-induced apoptosis, suggesting that inhibitors of these kinases may be useful therapeutically. Pretreatment with dasatinib, a broad spectrum tyrosine kinase inhibitor, blocked phosphorylation of PKCδ at both Tyr-64 and Tyr-155. Expression of “gate-keeper” mutants of c-Abl or c-Src that are active in the presence of dasatinib restored phosphorylation of PKCδ at Tyr-155 and Tyr-64, respectively. Imatinib, a c-Abl-selective inhibitor, also specifically blocked PKCδ Tyr-155 phosphorylation. Dasatinib and imatinib both blocked binding of PKCδ to importin-α and nuclear import, demonstrating that tyrosine kinase inhibitors can inhibit nuclear accumulation of PKCδ. Likewise, pretreatment with dasatinib also suppressed etoposide and radiation induced apoptosis in vitro. In vivo, pre-treatment of mice with dasatinib blocked radiation-induced apoptosis in the salivary gland by >60%. These data suggest that tyrosine kinase inhibitors may be useful prophylactically for protection of nontumor tissues in patients undergoing radiotherapy of the head and neck.

Radiation therapy for head and neck cancer can result in extensive damage to normal adjacent tissues such as the salivary gland and oral mucosa. We have shown previously that tyrosine phosphorylation at Tyr-64 and Tyr-155 activates PKC␦ in response to apoptotic stimuli by facilitating its nuclear import. Here we have identified the tyrosine kinases that mediate activation of PKC␦ in apoptotic cells and have explored the use of tyrosine kinase inhibitors for suppression of irradiation-induced apoptosis. We identify the damage-inducible kinase, c-Abl, as the PKC␦ Tyr-155 kinase and c-Src as the Tyr-64 kinase. Depletion of c-Abl or c-Src with shRNA decreased irradiation-and etoposide-induced apoptosis, suggesting that inhibitors of these kinases may be useful therapeutically. Pretreatment with dasatinib, a broad spectrum tyrosine kinase inhibitor, blocked phosphorylation of PKC␦ at both Tyr-64 and Tyr-155. Expression of "gate-keeper" mutants of c-Abl or c-Src that are active in the presence of dasatinib restored phosphorylation of PKC␦ at Tyr-155 and Tyr-64, respectively. Imatinib, a c-Abl-selective inhibitor, also specifically blocked PKC␦ Tyr-155 phosphorylation. Dasatinib and imatinib both blocked binding of PKC␦ to importin-␣ and nuclear import, demonstrating that tyrosine kinase inhibitors can inhibit nuclear accumulation of PKC␦. Likewise, pretreatment with dasatinib also suppressed etoposide and radiation induced apoptosis in vitro. In vivo, pre-treatment of mice with dasatinib blocked radiationinduced apoptosis in the salivary gland by >60%. These data suggest that tyrosine kinase inhibitors may be useful prophylactically for protection of nontumor tissues in patients undergoing radiotherapy of the head and neck.
Ionizing radiation (IR) 2 therapy in patients with head and neck cancer can result in moderate to severe damage to the salivary glands, resulting in loss of saliva, chronic oral infections, and severe discomfort (1). Other nontumor tissues in the oral cavity, such as the taste buds and oral mucosa, can also be unintended targets of IR damage; however, damage to these tissues generally resolves in the months after therapy (2). In contrast, damage to the salivary gland is permanent because it primarily affects salivary acinar cells which are postmitotic (1). Protein kinase C␦ (PKC␦) is essential for the apoptotic response of nontransformed cells to cell damaging agents (3)(4)(5)(6)(7), and our studies have defined PKC␦ as a critical regulator of IR-induced apoptosis in salivary epithelial cells (8 -13). Salivary epithelial cells from PKC␦ Ϫ/Ϫ mice are resistant to multiple apoptotic stimuli in vitro, whereas PKC␦ Ϫ/Ϫ mice are protected from IRinduced damage to the salivary gland and thymus in vivo and have a delay in mammary gland involution, a process driven by apoptosis (8, 14 -16).
PKC␦ is ubiquitously expressed and regulates a variety of cell functions in addition to apoptosis, including cell survival, migration, and proliferation (17). The ability of PKC␦ to control diverse cellular functions is due in part to tight regulation of its subcellular localization (17)(18)(19). In resting cells PKC␦ largely resides in the cytoplasm; however, upon DNA damage a series of highly regulated events results in its nuclear import and activation of downstream apoptotic pathways (9 -12). We reported previously that tyrosine phosphorylation of PKC␦ is rate-limiting for this process because phosphorylation at Tyr-64 and Tyr-155 results in a conformational change that facilitates importin-␣ binding to a C-terminal nuclear localization signal and nuclear import (10 -12). Candidate tyrosine kinases for phosphorylation of PKC␦ include c-Abl, which plays a prominent role in DNA repair, especially double-stranded break repair induced by DNA-damaging agents, and members of the Src family kinases (SFKs), known to control proliferation and cell migration (20 -24).
In our current studies we have identified the tyrosine kinases that mediate activation of PKC␦ in apoptotic cells and have * This work was supported, in whole or in part, by National Institutes of Health explored the use of TKIs (tyrosine kinase inhibitors) for protection of the salivary gland in patients undergoing radiotherapy for head and neck cancer. We show that phosphorylation of PKC␦ at Tyr-64 and Tyr-155, nuclear accumulation of PKC␦, and apoptosis can be specifically inhibited by pretreatment with TKIs. Our studies suggest that suppression of tyrosine phosphorylation of PKC␦ with TKIs may be a useful therapeutic strategy for protection of salivary gland function in patients undergoing head and neck irradiation.
Fluorescent Microscopy-ParC5 cells were grown on glass coverslips and transfected with pGFP-PKC␦. Following treatment with H 2 O 2 , coverslips were first rinsed with 1ϫ PBS (3 ϫ 10 min) then fixed with 2% paraformaldehyde for 15 min. Coverslips containing fixed cells were mounted on slides using Vectashield with DAPI mounting medium (H-1200; Vector Laboratories, Burlingame, CA). Subcellular localization of GFP-PKC␦ was analyzed by fluorescent microscopy. For quantification of GFP-PKC␦ localization, fixed cells were scanned using Olympus hardware and software (Center Valley, PA), and nuclear localization was quantified as the percentage of total cells with predominantly nuclear localized GFP. More than 200 cells were counted for each variable per experiment.
Analysis of Apoptosis in Vivo-C57BL/6 female mice were purchased from Jackson Laboratories (Bar Harbor, ME). Animals were maintained at the University of Colorado, Anschutz Medical Campus, in accordance with Laboratory Animal Care guidelines and protocols and with approval of the University of Colorado Denver Institutional Animal Use and Care Committee. Six-to 8-week-old female mice were left untreated or pretreated with dasatinib (20 mg/kg) via oral gavage 1 h prior to irradiation. Mice were anesthetized as described, and the head and neck region was irradiated using a cesium-137 source, while the remainder of the body was shielded with lead (8). Three h after irradiation, mice treated with dasatinib received a second dose of dasatinib (20 mg/kg). Mice were sacrificed 24 h following irradiation, and salivary glands were removed, fixed in 10% neutral buffered formalin, and embedded in paraffin for immunohistochemistry. Five-m sections were cut from the paraffin-embedded tissue for immunohistochemistry for detection of activated caspase-3 and counterstained with hematoxylin. For quantification of caspase-3 staining, sections were scanned Olympus hardware and software. Active caspase-3-positive cells in five random 40ϫ fields were quantified for each mouse (n ϭ 7 mice per condition). The data are expressed as the percentage caspase-3-positive cells/total epithelial cells.

c-Abl and c-Src Phosphorylate PKC␦ at Tyr-155 and Tyr-64,
Respectively-We have shown previously that PKC␦ is phosphorylated at Tyr-64 and Tyr-155 in response to H 2 O 2 and DNA-damaging agents and that tyrosine phosphorylation of PKC␦ is required for nuclear translocation and activation of its pro-apoptotic function (Refs. 11, 12 and Fig. 1A). Because tyrosine phosphorylation of PKC␦ may be a therapeutic target for regulation of apoptosis, we sought to identify the tyrosine kinases that phosphorylate PKC␦ at these sites. Both SFKs and c-Abl have been shown previously to phosphorylate PKC␦, and c-Src has been identified as the PKC␦ Tyr-311 kinase (21-23, 26, 27). To investigate the contribution of these tyrosine kinases to activation and nuclear import of PKC␦ following apoptotic signals, we treated salivary gland acinar cells (ParC5) with H 2 O 2 and assayed activation of c-Abl and SFKs using antibodies that specifically recognize the activated form of the tyrosine kinase. As seen in Fig. 1B, activation of both c-Abl and SFKs is evident by 5 min and persists for at least 40 min. Notably, tyrosine kinase activation parallels the kinetics of phosphorylation of PKC␦ at Tyr-64 and Tyr-155 (Ref. 12 and Fig. 2, A and C).
To demonstrate a specific role for c-Abl and c-Src in PKC␦dependent apoptosis we asked whether depletion of either tyrosine kinase with shRNA blocks phosphorylation of PKC␦ at Tyr-64 or Tyr-155. In ParC5 cells stably expressing a nontar-geting shRNA, treatment with H 2 O 2 results in phosphorylation of PKC␦ on Tyr-64, Tyr-155, and Tyr-311, whereas in cells expressing shRNA to c-Src, phosphorylation of Tyr-64 and Tyr-311, but not Tyr-155, is dramatically reduced (Fig. 1C). This confirms previous studies that have identified Tyr-311 as a c-Src site and suggests that Tyr-64 is also phosphorylated by c-Src (28). In contrast, depletion of c-Abl with shRNA leads to a significant reduction in phosphorylation of Tyr-155 (Fig. 1D). Unexpectedly, phosphorylation of PKC␦ at Tyr-64 was also reduced in cells depleted of c-Abl (Fig. 1D). Although it is possible that both c-Src and c-Abl can phosphorylate PKC␦ on Tyr-64, alternatively, phosphorylation at Tyr-64 by c-Src may be dependent upon prior phosphorylation of Tyr-155 by c-Abl. To test this directly, we transfected ParC5 cells with pGFP-PKC␦WT, or pGFP-PKC␦Y155F, a mutant that cannot be phosphorylated at Tyr-155, and assayed phosphorylation of PKC␦ at Tyr-64 ( Fig. 2A). In cells treated with H 2 O 2 , phosphorylation of PKC␦ at Tyr-64 was diminished and delayed in the context of the PKC␦ Y155F mutant ( Fig. 2A), suggesting that under these conditions, phosphorylation at Tyr-155 facilitates but is not required for phosphorylation of PKC␦ at Tyr-64.  Whole cell lysates were separated by SDS-PAGE and probed using phospho-specific antibodies against PKC␦ pY64, pY155, and pY311. To determine loading, membranes were stripped and probed for total PKC␦ and actin. B, ParC5 cells were treated with 5 mM H 2 O 2 for the indicated times. Whole cell lysates were resolved by SDS-PAGE and probed using a phospho-specific antibody against the c-Abl activation site (pY412) or the conserved SFK activation site (pY416). Membranes were stripped and re-probed for total c-Abl and c-Src. C and D, ParC5 cells stably expressing either an nontargeting shRNA or two unique shRNAs against c-Src (C) or c-Abl (D) were treated with 5 mM H 2 O 2 for 10 min. Whole cell lysates were resolved using SDS-PAGE and analyzed for PKC␦ pY64, pY155, and pY311. Blots were stripped and probed for total PKC␦ and actin. Efficiency and specificity of knockdown were determined by probing for c-Src and c-Abl. In C an asterisk denotes the band representing PKC␦ pY155. MW, molecular mass. However, in cells treated with etoposide, phosphorylation of the PKC␦ Y155F mutant at Tyr-64 was not detectable (Fig. 2B). In a reciprocal experiment where cells were transfected with pGFP-PKC␦Y64F and probed for phosphorylation of Tyr-155, no decrease in pY155 was detected in cells treated with either agent (Fig. 2, C and D). These data support a model in which phosphorylation of PKC␦ at Tyr-155 by c-Abl increases availability of the Tyr-64 site for phosphorylation by c-Src, presumably through a conformational change in the kinase which exposes the Tyr-64 site. The dependence upon phosphorylation at PKC␦ Tyr-155 for phosphorylation at Tyr-64 is much more apparent in the context of etoposide than H 2 O 2 . Notably, H 2 O 2 alone can induce oxidation-related conformational changes in proteins containing C1 domains. Because Tyr-155 in PKC␦ is adjacent to the C1 domain, oxidation by H 2 O 2 could result in some exposure and phosphorylation of Tyr-64 even in the absence of phosphorylation of PKC␦ at Tyr-155 (28).
Tyrosine Kinase Inhibitors Block Phosphorylation of PKC␦ at Tyr-64 and Tyr-155-TKIs encompass a large family of drugs that are used clinically for the treatment of neoplastic diseases (27). Our studies suggest that these drugs may also be useful for protection of nontumor tissue in patients undergoing IR treatment. To address this, we first explored whether TKIs could suppress tyrosine phosphorylation of PKC␦. ParC5 cells were treated with H 2 O 2 alone or following pretreatment with dasatinib, a broad spectrum TKI that inhibits both c-Abl and SFKs (29,30). Treatment with dasatinib plus H 2 O 2 suppresses acti-vation of c-Src and c-Abl dramatically and blocks phosphorylation of PKC␦ at Tyr-64 and Tyr-155, as well as the previously described c-Src site, Tyr-311 ( Fig. 3A and Ref. 26). Notably, we consistently observed a reduction (average decrease 45%) in total Src protein with Src activation (see Fig. 3, A, B, C, right, and D).
To confirm that inhibition of tyrosine phosphorylation indeed results from suppression of c-Src and c-Abl activation in dasatinib-treated cells, we repeated this experiment in cells transfected with plasmids encoding gatekeeper mutants for c-Src (T341I) (Fig. 3C) or c-Abl (T315I) (Fig. 3B). These mutants are resistant to inhibition by dasatinib (31 (Fig. 3C). Phosphorylation at Tyr-155 was not rescued by c-Src T341I, but was rescued by expression of the c-Abl gatekeeper mutant, c-Abl T315I (Fig. 3B)   Whole cell lysates were resolved using SDS-PAGE and analyzed for PKC␦ pY64, pY155, and pY311. Inhibition of c-Src and c-Abl was determined by probing for their respective activation sites pY416 (c-Src) and pY412 (c-Abl). Blots were stripped and probed for total PKC␦, total c-Src, total c-Abl, and actin. B and C, 293T cells were transfected with either wild type c-Src (B), wild type c-Abl (C), or the gatekeeper mutation for c-Src T341I (B) or c-Abl T315I (C). Transfected cells were treated as in A. Whole cell lysates were resolved using SDS-PAGE and analyzed for PKC␦ pY64, pY155, and pY311. For B and C, an asterisk denotes the band representing PKC␦ pY155. Blots were stripped and probed for total PKC␦, total c-Src, total c-Abl, and actin. For D, an asterisk distinguishes the band representing c-Abl pY412 from a lower background band.
Abl T341I completely restored phosphorylation at Tyr-155 in cells treated with H 2 O 2 plus dasatinib (Fig. 3B). Interestingly, expression of either gatekeeper mutant resulted in an increase in basal phosphorylation at their respective activation sites (pY412 for c-Abl and pY416 for c-Src) (Fig. 3, B, 3C, and Ref. 31). Finally, we show that in ParC5 cells pretreated with the c-Abl-selective inhibitor imatinib, phosphorylation of PKC␦ at Tyr-155 is also abolished ( Fig. 3D and Ref. 30). The dose of imatinib used had very minimal effects on c-Src activation by H 2 O 2 and did not inhibit phosphorylation of the c-Src site, Tyr-311. However, imatinib did reduce phosphorylation of Tyr-64, again consistent with our previous data that suggests co-operativity between phosphorylation at Tyr-64 and Tyr-155.
Tyrosine Kinase Inhibition Suppresses Nuclear Localization of PKC␦-Phosphorylation of PKC␦ on Tyr-64 and Tyr-155 regulates its nuclear import in response to apoptotic agents by facilitating interaction with importin-␣ (12). Based on our observation that treatment with TKIs blocks Tyr-64 and Tyr-155 phosphorylation, we predicted that TKIs would prevent nuclear translocation of PKC␦. To address this, we asked whether treatment with dasatinib results in exclusion of PKC␦ from the nucleus in cells treated with H 2 O 2 . ParC5 cells were transfected with pGFP-PKC␦ and treated with H 2 O 2 alone or in combination with dasatinib, and cells with nuclear PKC␦ were quantified. Nuclear localization of GFP-PKC␦ was observed in 27% of ParC5 cells treated with H 2 O 2 for 30 min and 56% of cells treated for 60 min. Nuclear accumulation of GFP-PKC␦ was substantially reduced in cells pretreated with dasatinib prior to the addition of H 2 O 2 (15% at 30 min) and (25% at 60 min) (Fig. 4A). We have shown previously that tyrosine phosphorylation results in a conformational change in the kinase, exposing the binding site for importin-␣ (12). To determine whether dasatinib suppression of nuclear accumulation of GFP-PKC␦ is because of reduced importin-␣ binding, GFP-PKC␦ was immunoprecipitated and probed for bound importin-␣. Whereas treatment with H 2 O 2 increased importin-␣ binding to PKC␦, this was dramatically reduced in cells pretreated with dasatinib (Fig. 4B, left). Likewise, pretreatment with the c-Abl-selective inhibitor, imatinib, also substantially reduced H 2 O 2 induced binding of importin-␣ (Fig. 4B, right).
The studies above suggest that PKC␦-dependent apoptotic signaling, and possibly apoptosis, can be targeted by inhibition of specific tyrosine kinases through the use of specific TKIs. To explore this, we first utilized ParC5 cells that express two unique shRNAs against c-Src, c-Abl, or a scrambled control shRNA. ParC5 cells depleted of c-Src or c-Abl were treated with IR or the DNA damaging agent etoposide, and activation of caspase-3 was assayed (Fig. 5, A-D). When ParC5 cells depleted of c-Src or c-Abl were treated with etoposide there was up to a 50% reduction in caspase-3 activation compared with cells expressing the scrambled control shRNA (Fig. 5, A and B). Similar results were seen when c-Src-or c-Abl-depleted cells were exposed to IR (Fig. 5, C and D). These studies confirm and  extend previous work from our laboratory by demonstrating that c-Src and c-Abl play an essential role in DNA damageinduced apoptosis through phosphorylation of PKC␦ at Tyr-64 and Tyr-155 (11,12).
To determine the potential use of TKIs to protect against salivary gland damage, we asked whether treatment with dasatinib could suppress apoptosis either in vitro or in mice exposed to head and neck IR. In vitro pretreatment with dasatinib suppressed apoptosis by 80% in cells treated with 10 Gy ␥-irradiation (Fig. 5E). For the in vivo analysis, dasatinib was administered via oral gavage 60 min before and again 3 h after exposure to IR. Dasatinib treatment reduced IR-induced apoptosis by Ͼ60% in the parotid salivary gland, as measured by analysis of caspase-3 activation (Fig. 6). These data demonstrate a potential novel use for TKIs as a prophylactic strategy for protection of the salivary gland and possibly other oral tissues in head and neck cancer patients receiving radiation therapy.

DISCUSSION
Radiation therapy for head and neck cancer patients can result in extensive and permanent damage to adjacent healthy tissues including the salivary gland. Therapeutics designed to protect the salivary gland, such as the free radical scavenger amifostine, have limited efficacy and significant toxicity (32).
Our previous studies have identified tyrosine phosphorylation of PKC␦ as a rate-limited step in IR-induced apoptosis in the salivary gland (11). Here we show that TKIs effective against c-Src and c-Abl are able to block multiple key regulatory steps necessary for PKC␦ nuclear localization leading to suppression of apoptosis both in vitro and in vivo. Our data suggest that some TKIs, currently used in the clinic for treatment of cancer, may also be useful for protection of nontumor tissues in patients undergoing radiotherapy of the head and neck.
Our studies demonstrate a role for the tyrosine kinases c-Src and c-Abl in tyrosine phosphorylation and in turn activation of PKC␦ in response to apoptotic stimuli. Specifically, we define c-Abl as the Tyr-155 kinase. Phosphorylation of PKC␦ at Tyr-155 is blocked by the c-Abl inhibitors dasatinib and imatinib and by c-Abl depletion using shRNA. Additionally, expression of the c-Abl gatekeeper mutant (T315I) can specifically restore phosphorylation at Tyr-155 in the presence of dasatinib. Our data support and extend previous studies that have reported a functional relationship between c-Abl and PKC␦ in response to genotoxic and oxidative stress (33)(34)(35). Both Yuan et al. (34) and Sun et al. (33) have shown that IR and H 2 O 2 induce phosphorylation of PKC␦ by c-Abl, although the specific tyrosine phosphorylated was not identified. Intriguingly, a recent report showed that hyperglycemia induced apoptosis of neural progenitor cells occurs through a PKC␦⅐c-Abl-dependent mechanism that involves tyrosine phosphorylation of PKC␦ and nuclear translocation of the PKC␦⅐c-Abl complex (36). Previous reports have demonstrated a role for members of the Src family of non-receptor tyrosine kinases in the regulation of PKC␦ (21,37,38). For instance, phosphorylation of PKC␦ by c-Src has been shown to control its degradation and activation (23,39,40). c-Src and PKC␦ also mediate signal transduction through growth factor receptors, including platelet-derived growth factor receptor, extracellular growth factor receptor, and the insulin receptor (24,41). In addition, c-Src dependent tyrosine phosphorylation has been shown to regulate the proapoptotic function of PKC␦ in neuronal cells and in response to ceramide and chemotherapeutic drugs (21,38,42). Our studies strongly support a role for c-Src as the PKC␦ Tyr-64 kinase. We show that stable depletion of c-Src using shRNA blocks Tyr-64 phosphorylation and that expression of the Src T341I gatekeeper mutant specifically restores Tyr-64 phosphorylation in the presence of dasatinib (Figs. 1C and 3B). Unexpectedly, inhibition or depletion of c-Abl can also partially reduce phosphorylation at Tyr-64 independent of suppression of Src. This is explained at least in part by our findings that Tyr-155 phosphorylation by c-Abl facilitates phosphorylation of Tyr-64. However, c-Abl may also be able to directly phosphorylate the Tyr-64 site as some phosphorylation of Tyr-64 is recovered in dasatinib-treated cells that express the c-Abl gatekeeper mutation, under conditions where c-Src activity is inhibited (Fig.  3B).
We have shown previously that phosphorylation of Tyr-64 and Tyr-155 promotes a conformational change in PKC␦ leading to its association with importin-␣ and subsequent nuclear translocation (12). Our current data suggest a hierarchical relationship between Tyr-155 and Tyr-64, where phosphorylation of Tyr-155 by c-Abl serves as a permissive signal for phosphorylation of Tyr-64. These data support a model in which phosphorylation of PKC␦ at Tyr-155 presumably leads to a structural change that makes Tyr-64 more accessible to phosphorylation by c-Src. Notably, phosphorylation of the PKC␦ Y155F mutant at Tyr-64 was completely lost in cells treated with etoposide, but only reduced in cells treated with H 2 O 2 (Fig.  2, A and B). This is consistent with a partial loss of protein structure in the presence of H 2 O 2 but not etoposide (28,43). Sequential phosphorylation of PKC␦ by c-Abl and c-Src is likely to be critical to cellular homeostasis. Because c-Abl is a damage-induced tyrosine kinase, this assures that pro-apoptotic signaling by PKC␦ is coordinated with activation of other cell death signals. Furthermore, this model explains how c-Src, which regulates many survival/proliferation signals, may also play a role in apoptosis. Based on this relationship, the kinase activity of c-Abl and/or c-Src provides a logical target for the disruption of the pro-apoptotic function of PKC␦.
TKIs have been developed against many kinases required for cancer cell proliferation, including members of the Src family and c-Abl (27). Here we explored the novel use of these inhibitors as prophylactic agents to prevent IR-induced cell death within the salivary gland. We show that pretreatment with dasatinib, which inhibits SFKs and c-Abl, is sufficient to suppress activation and nuclear translocation of PKC␦. Further, a 20-mg/kg dose in our mouse model equates to a serum concentration that is easily obtained in patient populations (44,45). Expectedly, pretreatment with the c-Abl-selective inhibitor imatinib also blocked nuclear translocation, as evidenced by reduced binding of importin-␣ to PKC␦ following H 2 O 2 treatment. Importantly, treatment of mice with dasatinib significantly reduced IR-induced salivary gland apoptosis. Taken together our in vitro and in in vivo data suggest that TKIs can offer radioprotection within the salivary gland by inhibiting the activation of PKC␦. These findings support our previous studies which showed that IR-induced apoptosis in the salivary gland is significantly reduced in PKC␦ Ϫ/Ϫ mice (46) and recent studies by Ovitt and co-workers which demonstrate that depletion of PKC␦ within the mouse salivary gland using nanoparticle delivered PKC␦ siRNA protects against IR-induced loss of salivary gland function (47). Pabla et al. have also recently shown that PKC␦ inhibition leads to a reduction in nephrotoxicity caused by cisplatin with no effect on the efficaciousness of cisplatin (38).
Our findings provide a rationale for future clinical trials to investigate TKIs as a radioprotective therapy to preserve salivary function in patients being treated for head and neck cancer. In the context of IR treatment, our studies suggest that short term dosing with a TKI may be sufficient. Most patients receive fractionated IR over the course of a few months, thus treatment with TKIs would likely be limited to that time frame. An important concern with all radioprotective treatments is that they do not promote tumor growth or hamper tumor therapy. Although few TKIs have been tested in patients with head and neck cancer, phase I/II clinical trials with dasatinib showed no enhancement of tumor growth or progression (48). At the same time this study failed to demonstrate activity of dasatinib as a single agent against head and neck cancer. However, a report by Lin et al. shows that IR can be combined with dasatinib to further sensitize head and neck squamous cell carcinoma cells (49). Therefore, it can be suggested that dasatinib will protect salivary glands while if anything promoting apoptosis of tumor tissue in patients receiving IR treatment for head and neck cancer. Finally, our studies may have more far reaching implications for protection of other nontumor tissues in patients undergoing IR or some types of chemotherapy.