PKC is an indispensable factor in promoting environmental toxin chromium-mediated transformation and drug resistance

Cr(VI) stimulates and sustains PKC activity

PKC, as an important signal transducer, participates in regulating cell proliferation, migration and apoptosis [28–30]. However, the role of PKC in Cr(VI)-mediated carcinogenesis remains unclear, regardless of the studies showing the involvement of several mitogenic pathways (such as Src/Ras, JNJ/p38) in this metal toxin-induced actions. Therefore, PKC expression in human bronchial epithelial cells and keratinocytes with or without transient (2 h) or chronic exposure (<2 h) to Cr(VI) was first tested by immunoblotting (Figure 1A). The expression levels of this kinase in the cells were not altered by Cr(VI) treatment at different testing times. PKC activity was then analyzed after transient or chronic Cr(VI) exposure (Figure 1B). Phorbol 12-myristate 13-acetate (PMA, a PKC stimulator) was used as a positive control, and strongly stimulated PKC activity. The transient Cr(VI) treatment moderately upregulated PKC activity, which was slightly declined, but sustained in response to the prolonged exposure. AP-1 (a transcription factor) is downstream of PKC and composed of homo- or heterodimers of c-Jun, c-Fos or ATF [43]. To further determine PKC activation, the phosphorylation or activation status of c-Jun in the cells was examined following Cr(VI) exposure at different times (Figure 1C). c-Jun was phosphorylated in the cells received Cr(VI) treatment for 2 h, which was attenuated, but maintained at a moderate level following the chronic exposure (Figure 1C, left panels). The folds of increased, phosphorylated c-Jun were also measured by the phosphor-imager (Figure 1C, right panel).

Figure 1. Activation of PKC by Cr(VI) in BEAS-2B cells and Keratinocytes. (A) Cells were treated with Cr(VI) (2.0 uM) for different times. Subsequently, PKC expression was tested by immunoblotting. Actin was the loading control. (B) After Cr(VI) treatment for 2 h or 1 month (1M), PKC activity in the cells was analyzed by PKC kinase activity assay. PMA (0.5 μM) stimulation serves as positive control. The error bars represent standard deviation (SD) (n = 5, p < 0.01). (C) Cells were treated as described above and lysates were prepared. Subsequently, the phosphorylated c-Jun was examined by immunoblotting (left panels). The folds of the proteins induced were also measured and plotted (right panel). Actin was the loading control.

PKC activation by Cr(VI) was independent of the Src/Ras signaling

Src/Ras signaling pathway was suggested to be cooperated with PKC in regulating different cellular activities by mitogenic stimulation [28–30]. It was not clear about if and how PKC and Src/Ras pathways were interacted in response to Cr(VI) exposure. To test this, PKC activity, after Cr(VI) treatment for different times, was analyzed in the presence and absence of GO6976 (a PKC inhibitor), PP1 [4-amino-5-(4-methyphenyl)-7-(t-butyl) pyrazolo (3.4-d)-pyrimidine, a Src inhibitor] or FTI (farnesyltransferase, a Ras inhibitor) (Figure 2A). GO6976 inhibited the upregulation of PKC induced by transient or chronic Cr(VI) exposure, as expected. However, The addition of PP1 or FTI respectively, played no role in Cr(VI)-mediated PKC upregulation in the cells.

Figure 2. PKC activation by Cr(VI) was independent of Src/Ras signaling. (A) BEAS-2B cells and keratinocytes were treated with Cr(VI) for 2 h or 1 month in the presence or absence of GO6976 (1.0 μM), FTI (0.25 μM) or PP1 (0.15 μM). Afterwards, lysates were prepared for PKC activity assay. The error bars were SD (n = 5, p < 0.05). (B) BEAS-2B cells were exposed to Cr(VI) for 2 h or 24 h in the presence or absence of GO6976 and then assayed for the expressions of the phosphorylated Src by immunoblotting and active Ras using active Ras pull-down kit. Src and actin were the loading controls. (C) Cells were treated as described above and levels of ROS were then analyzed. The error bars were SD (n = 5, p < 0.01).

To further determine the involvement of PKC in Cr(VI), via PKC, activated Src or Ras signaling, BEAS-2B cells were treated with Cr(VI) for 2 h or 1 day in the presence or absence of GO6976 and afterwards the activation of Src or Ras was analyzed (Figure 2B). Src was phosphorylated and active Ras was detected in the cells after Cr(VI) exposure at the times tested, suggesting that the inhibition of PKC did not affect Cr(VI)-mediated activation of these two signal transducers.

Cr(VI) was shown to via activating the Src/Ras signaling, aberrantly increased ROS levels and further disrupted redox balance in normal cells [27]. To test the involvement of PKC in Cr(VI)-mediated perturbation of redox signaling, the amounts of ROS was analyzed (Figure 2C). Two hours of Cr(VI) exposure strongly augmented ROS in BEAS-2B cells and keratinocytes, which was slightly reduced, but sustained by the chronic exposure. The addition of PP1 or FTI, but not of GO6976, suppressed Cr(VI)-mediated ROS upregulation. Thus, the results indicate that the increase of ROS by Cr(VI) is regulated by Src/Ras signaling and independent of PKC.

PKC, via upregulating Bcl-2, desensitizes Cr(VI)-treated cells to doxorubicin-mediated cytotoxicity

Cr(VI)-initiated cancer cells, via desensitizing the intrinsic cell machinery, often develop drug resistance, which is a major challenge for chemotherapies [43, 44]. In order to explore the mechanisms, the sensitivity of the cells transiently or chronically treated with Cr(VI) to apoptosis induced by doxorubicin (a chemo-drug) was examined (Figure 3A). After being exposed to Cr(VI) for either 2 h or one month, BEAS-2B cells and keratinocytes were then treated with doxorubicin for 48 h and the percentages of the cells with fragmented DNA were measured. Approximately 40% of the cells treated with doxorubicin alone or transiently exposed (2 h) to Cr(VI) underwent apoptosis. In comparison, less than 15% of the cells received chronic Cr(VI) exposure became apoptotic after doxorubicin treatment, indicating that a long-period exposure of Cr(VI) impairs the intrinsic cell death machinery.

Figure 3. Upregulation of Bcl-2 by Cr(VI) was via increasing its stability and depends upon PKC. (A) Cells were exposed to Cr(VI) for 2 h and 1 month and their sensitivities to doxorubicin (5 μM)-induced apoptosis were analyzed by DNA fragmentation assay. The error bars were SD (n = 5, p < 0.01). (B) BEAS-2B cells were transiently or persistently exposed to Cr(VI) prior to doxorubicin treatment, and then examined for Bcl-2 expression. Actin was the loading control. (C) After the treatments as described above, Bax expression was analyzed. Actin was the loading control. (D) After adding GO6976, Bcl-2 expression in the cells received the treatments as described above was tested by immunoblotting. Actin was the loading control. (E) Cells chronically exposed to Cr(VI) were treated with doxorubicin or co-treated with GO6976, prior adding CHX (0.5 μM). Bcl-2 expression was then analyzed at different time points of CHX block. Actin was the loading control. (F) Cells chronically exposed to Cr(VI) were treated with doxorubicin. Afterwards, actinomycin D (ATC, 1.0 nM) was added and expression levels of bcl-2 gene were analyzed at different time points of ATC block by real time PCR. The error bars were SD (n = 5, p < 0.05).

Bcl-2 is a pro-survival factor and PKC was shown to be involved in the regulation of Bcl-2 function [41, 42, 45]. To test if and how these two factors might be involved in the resistance to doxorubicin in chronic Cr(VI)-treated cells, the expression of Bcl-2 in Cr(VI)-treated BEAS-2B cells, after doxorubicin treatment, was analyzed by immunoblotting (Figure 3B). A baseline expression of Bcl-2 was detected in untreated cells, which was augmented in the cells treated with doxorubicin alone and co-exposed to transient or chronic Cr(VI). Bax is a pro-apoptotic factor and able to interact with Bcl-2 [46]. The expression of Bax after the same treatments described above was then tested (Figure 3C). The level of Bax in BEAS-2B cells with or without 2 h of Cr(VI) exposure was increased, which did not occur in the cells chronically exposed to Cr(VI). The data indicates that Bax is induced by doxorubicin and chronic Cr(VI) exposure appears hindering Bax induction, which may facilitate in strengthening Bcl-2 anti-apoptotic function.

The involvement of PKC in the promotion of Bcl-2 function during various apoptotic processes has been reported [41–44]. To test the importance of the cooperation of PKC and Bcl-2, BEAS-2B cells and keratinocytes chronically exposed to Cr(VI) were treated with doxorubicin in the presence or absence of GO6976 (Figure 3D). Subsequently immunoblotting for Bcl-2 expression was performed. Again, Bcl-2 expression in BEAS-2B cells or keratinocytes chronically exposed to Cr(VI) was induced by doxorubicin, which was suppressed by GO6976, indicating the linear relationship between these two molecules and also suggesting that this anti-apoptotic factor functions downstream of PKC.

Protein stability is one of the elements for the regulation of protein expressions [47]. Therefore, we tested if the upregulation of Bcl-2 expression by Cr(VI), in response to doxorubicin, was due to the influence of PKC on its degradation process (Figure 3E). The chronic Cr(VI)-exposed BEAS-2B cells were treated with either GO6976 or FTI, prior to adding doxorubicin following blocking protein synthesis by cycloheximide (CHX). Subsequently, Bcl-2 expression levels were examined at different times of CHX block. This anti-apoptotic factor in chronic Cr(VI)-treated cells was much more stable after doxorubicin treatment than that in untreated cells. After the suppression of PKC by GO6976, the kinetics of Bcl-2 degradation in chronic Cr(VI)-treated cells, after doxorubicin treatment, became similar as that of untreated cells. In the presence of FTI, the attenuation of Bcl-2 degradation in the cell co-exposed to chronic Cr(VI) treatment and doxorubicin was unchanged. Next, the stability of bcl-2 gene in chronic Cr(VI)-treated cells, after the addition doxorubicin, was also examined, using Real-Time PCR analysis (Figure 3F). After the addition of actinomycin D (ATC) to block gene transcription, the levels of bcl-2 in the treated- or untreated cells were similar, indicating that the increase of Bcl-2 expression by Cr(VI) is not regulated at the transcriptional level. Taken together, the results suggest that PKC, via influencing protein degradation process, stabilizes Bcl-2 in chronic Cr(VI) exposed cells, which is in a Ras independent fashion.

PKC cooperates with Ras pathway, to promoted Cr(VI)-mediated drug resistance and cell growth

To further test the role of PKC or Ras in our experimental setting, BEAS-2B cells and keratinocytes, after being chronically exposed to Cr(VI), were treated with doxorubicin in the presence or absence of GO6976, FTI or both for 48 h and then subjected DNA fragmentation analysis (Figure 4A). The cells chronically exposed to Cr(VI) were much less sensitive to doxorubicin for the induction of apoptosis than the control cells. The inhibition of PKC or Ras, respectively, partially restored the sensitivity of chronic Cr(VI) treated cells (about 25%) to doxorubicin for the induction of apoptosis. When being co-treated with GO6976 and FTI, more than 40% of the cells chronically exposed to Cr(VI) were sensitized to doxorubicin-induced apoptosis. The similar results were obtained from Annexin V analysis (Figure 4B). The results suggest that PKC and Ras both take part in the regulation of Cr(VI)-mediated drug resistance, but function in the separate, but cooperative pathways.

Figure 4. PKC and Ras cooperate to promote Cr(VI)-mediated drug resistance and long-term survival. (A) Cells with or without chronic Cr(VI) exposure were received different treatments. Subsequently, the percentages of the cells with fragmented DNAs were measured. The error bars were SD (n = 5, p < 0.01). (B) After the same treatments as described above, the cells were subjected to Annexin V apoptotic assay. The error bars were SD (n = 5, p < 0.01). (C) After the treatments described above, colony formation assay was conducted. The plating efficiencies were measured (right panel) and colonies were imaged (right panels). The error bars were SD (n = 5, p < 0.05).

To further test the effect of PKC or Ras on long-term survival of chronic Cr(VI)-exposed cells in response to doxorubicin treatment, the clonogenicity assay was conducted (Figure 4C). BEAS-2B cells chronically exposed to Cr(VI) were treated with doxorubicin in the presence or absence of GO6976 or FTI and then grew for 15 days to allow forming colonies. The plating efficiencies showed that chronic Cr(VI) exposure enhanced colony formation capability of the cells in response to doxorubicin (Figure 4C, left panel). The inhibition of PKC or Ras respectively, partially reduced the ability of chronic Cr(VI)-treated cells to doxorubicin. In comparison, the co-suppression of these two pathways abolished the resistance to this chemo drug. The images of the colonies clearly indicated the effect of the suppression of PKC, Ras or both on the clonogenicity of the cells chronically exposed to Cr(VI), following doxorubicin treatment (Figure 4C, right panels). Thus, the data indicate that persistent Cr(VI) exposure, via activating PKC and Ras, not only promotes cell survival, but also drug resistance.

PKC and Ras are required for Cr(VI)-mediated transformation

To test the effect of chronic Cr(VI) exposure on cellular transformation, BEAS-2B cells and keratinocytes were grown in soft agar medium containing Cr(VI) in the presence or absence of FTI, GO6976 or both for 4.5 months and the numbers of colonies were then counted (Figure 5A). A few of untreated control BEAS-2B cells or keratinocytes were formed colonies (Figure 5A, left panel). In comparison, approximately 5% of chronic Cr(VI)-treated cells formed colonies in soft agar medium. The co-treatment of Cr(VI) with FTI or GO6976, respectively, partially suppressed this transformation process. In the presence of both GO6976 and FTI, this metal toxin became unable to transform the cells. The images showed the colonies from BEAS-2B cells received different treatments (Figure 5A, right panels). Furthermore, the expressions of active Ras, phosphorylated c-Jun or Bcl-2 in two chronic Cr(VI)-treated, transformed colonies were analyzed, using active Ras pull down assay or immunoblotting analysis (Figure 5B, left panels) and folds of the inductions of these factors were measured (Figure 5B, right panels). Consistently, active Ras, p-c-Jun and increased Bcl-2 were detected in Cr(VI)-transformed BEAS-2B cells. The results further support the notion that Ras and PKC signaling pathways are indispensable elements in Cr(VI)-mediated transformation.

Figure 5. Role of PKC or Src/Ras signaling in Cr(VI)-mediated transformation. (A) Cells were cultured in soft agar containing Cr(VI) or Cr(VI) plus PP1, FTI or GO6976, respectively for another 4.5 months. Untreated cells grown in soft agar medium served as the control. The numbers of the colonies were counted (left panel) and examples of colonies from the treatments were imaged (right panels). The error bars were SD (n = 5, p < 0.05). (B) Expression of active Ras, phosphorylated c-Jun and Bcl-2 in two Cr(VI) transformed colonies were examined by immunoblotting (left panels). The folds of the proteins induced were measured and then plotted (right panels). Actin was the loading control.

Cells and reagents

Human BEAS-2B lung epithelial cells and human keratinocytes were purchased from ATCC (Manassas, VA). The cells grew in RPMI 1640 supplemented with 10% heat-inactivated fetal bovine serum (Invitrogen, CA) in a 5% CO₂ humidified atmosphere at 37°C. The authentication of these cell lines is based on the information provided by the company. In addition, the cells were periodically authenticated by monitoring cell morphology and growth curves. The concentration of Cr(VI) used in this study is (2.0 μM). The transient Cr(VI) exposure represent the treatment for 2 h and more than 2 h of the exposures are stated in the text as the persistent or chronic treatment.

Cr(VI), GO6976, FTI, PP1, PMA and doxorubicin were purchased from Sigman (St Louis, MO, USA). Antibodies against p-Src, Src, β-actin, pan-PKC, Bax, p-c-Jun and c-Jun were purchased from Santa Cruz Biotechnology (Santa Cruz, CA, USA). Anti-Bcl-2 antibody was from Cell Signaling Technology (Danvers, MA, USA). The authentication information was provided by each company.

Immunoblotting analysis

After treatments, cells were harvested, washed with 1 x PBS twice and lysed in the lysis buffer (50 mM Tris-HCl, pH 8.0, 150 mM NaCl, 1% Triton-X114, 0.5% sodium deoxycholate, 0.1% sodium dodecyl sulfate, supplemented with protease and phosphatase inhibitor cocktails (Thermo Scientific, Waltham, MA, USA). Subsequently, cell lysates were separated by 10% SDS-PAGE gels (Thermo Scientific, Waltham, MA, USA) and transferred to Immobilon-P membrane (Millipore, Burlington, MA, USA). After blocked with TBS containing 5% of dried milk, membranes were probed with primary antibodies, followed by added ECL, anti-rabbit IgG, anti-mouse IgG or anti-goat IgG secondary antibody (1:5,000-10,000) and then imaged. All immunoblotting experiments were conducted at least twice.

PKC activity assay

PKC activity was tested using the kit from Abcam (Cambridge, MA, USA). According to the protocol provided by the company: after treatments, samples were added into each well of a 96 well-plate and then suspended with 10 μl of diluted ATP per well. After keeping at 30°C for 90 min on a rotating shaker, 40 μl of the antibody against corresponding phosphor-specific substrate was added into each well. Following 60 min of incubation, liquid in each well was aspirated and samples were washed with the wash buffer. Subsequently, a second antibody was added followed by the substrate. Finally, the stop solution was added and absorbance was read by a microplate reader. For each treatment, 5 wells were used and the experiments were repeated twice.

Active Ras pull-down assay

After treatments, cell lysates were collected and assayed by the active Ras pull-down and detection kit, as instructed by the manufacturer (Thermo Scientific, Waltham, MA, USA). Briefly, the GTP-form of Ras was pulled down by a GST-fusion protein with the Ras-binding domain (RBD) of Raf attached to glutathione agarose. The pull-down complexes were washed and separated on a 10% SDS-PAGE gel and immunoblotted with an anti-pan-Ras antibody. For loading controls, equal amounts of lysates as that for the assay were subjected to immunoblotting and then probed with anti- β-actin antibody. The assay was conducted at least two times.

ROS analysis

After treatments, cells were washed with ice-cold PBS and resuspended in 5 μg/ml of 2′, 7′-dichlorodihydrofluorescein diacetate (DCF) (Thermo Fisher Scientific, MA, USA). Following by incubation for 10 min at room temperature, the samples were immediately analyzed.

DNA fragmentation apoptotic assay

A flow cytometric analysis was performed using a FACScan (BD Biosciences, Franklin Lakes, NJ, USA). The data analysis was performed using the Cell-Fit software program (BD Biosciences). Cell-Fit receives data from the flow cytometer and provides real-time statistical analysis, computed at 1-s intervals, and also discriminates doublets or adjacent particles. Cells with sub-G₀-G₁ DNA contents after staining with propidium iodide were counted as apoptotic cells. In brief, after treatments, the cells were harvested and then fixed in 70% cold ethanol. Afterward, cells were stained with 0.01 μg/ml of propidium iodide containing 1.5 ng/ml of RNase. DNA contents of cells were then tested using a FACScan machine. The experiments were repeated three times to verify reproducibility.

Clonogenicity assay

Clonogenicity assay is to measure the ability of a cell to grow and form a colony. Cells (1000 cells/dish) with or without chronic Cr(VI) exposure were seeded into 10 cm diameter plates and allowed to adhere overnight. Cells then were treated with doxorubicin in the presence or absence of different inhibitors and grown for 15 days meanwhile the culture medium was changed every 3 days. Afterwards, cells were fixed with the fixation solution containing 1% methanol and 1% formaldehyde, and stained with 0.5% crystal violet. A cluster containing more than 60 cells was considered as a colony. The numbers of colonies were counted using an optical microscope and plating efficiencies were plotted as percentages of numbers of treated cells versus untreated cells. For each treatment, 5 dishes were used.

Soft agar assay

Cells (2500 cells/dish) were grown in soft agar medium in the presence or absence of Cr(VI) or Cr(VI) plus the inhibitors for 4.5 months. Meanwhile, 0.5 ml of fresh medium with or without Cr(VI) or Cr(VI) plus the inhibitors was added on the top of the soft agar every 4 days. At the end of experiments, plates were stained with crystal violet and colonies were counted under a dissecting microscope.

Statistical analysis

Statistical analysis was performed using a two-tailed Student’s t test for comparison of two groups or a one-way ANOVA analysis. Standard deviations are displayed in the figures. A p value < 0.05 was considered significant.