Tumor hypoxia, mostly resulting from poor perfusion and anemia, is one of the key factors involved in the development of cell clones with an aggressive and treatment-resistant phenotype, which leads to rapid progression and poor prognosis (Semenza,2001; Poellinger and Johnson,2004; Vaupel,2008). It has been reported that solid tumors in humans, even those less than 1 cm in diameter (i.e., at the limits of clinical detection), may have substantial hypoxic fractions (Bae et al.,1999). Tumor hypoxia has adverse effects on the outcomes of clinical radiotherapy and chemotherapy. Hypoxia causes adaptive changes in the transcription of a wide range of genes involved either in increasing the availability of oxygen to the tissues or in decreasing the cellular consumption of oxygen (Brahimi-Horn et al.,2001). In addition, hypoxia is known to directly and indirectly confer resistance to X- and γ-radiation and some chemotherapies, which results in treatment failure.
The early growth response 1 (Egr-1) gene, also known as NGF1-A, TIS8, Krox-24, and Zif268, is rapidly induced in response to a variety of stimuli, including growth factors, cytokines, hypoxia, physical forces, and tissue injury (Lemaire et al.,1988; Sukhatme et al.,1988; Khachigian,2006). EGR-1 is considered to be a central regulator of tumor cell proliferation, migration, and angiogenesis and has shown to be a promising candidate target for gene therapy in human astrocytomas (Mittelbronn et al.,2009). It has been reported that hypoxia can trigger Egr-1 accumulation (Banks et al.,2005), but the mechanisms by which Egr-1 affects tumor cell viability and proliferation during hypoxia are poorly understood.
Microtubules are a critical element in a wide variety of fundamental functions, including maintaining the cell shape, regulation of cell motility, and cell division (Hyams and Lloyd,1993; Margolis and Wilson,1998; Ling et al.,2002). Microtubule arrays are assembled into a bioriented spindle, which is responsible for aligning chromosomes during the metaphase and separation chromosomes during the anaphase. The progression of mitosis is arrested until the spindle has been correctly assembled during the metaphase through a regulatory system called the spindle checkpoint system. Microtubule-targeting cancer therapies interfere with mitotic spindle dynamics and block cells in mitosis by activating the mitotic checkpoint. Therefore, microtubule poisons such as vinca alkaloids and taxanes are useful therapeutic compounds for the treatment of human cancer (Rowinsky and Donehower,1991; Jordan et al.,1992). Tumor cells often have a defective spindle checkpoint system (Pihan et al.,2003) and attempt to progress through mitosis despite the presence of microtubule defects. This results in massive chromosomal segregation error and ultimately triggers the programmed cell death signals to kill the tumor cells (Ling et al.,2002).
In this study, we found that the expression of Egr-1 was increased when the tumor cells were exposed to hypoxia, and stabilized the microtubule network, which was confirmed by immunofluorescence. Knockdown of Egr-1 increases the sensitivity of the tumor cells to the microtubule poisons commonly used to treat the human cancer under hypoxic conditions.
MATERIALS AND METHODS
Cell Culture and Hypoxia Treatment
BEL-7402 cells (human hepatocellular carcinoma cell) were purchased from the China Center for Type Culture Collection and were cultured in Dulbecco's modified Eagle's medium (Gibco/BRL, Gaithersburg, MD) supplemented with 10% heat-inactivated fetal bovine serum. Media containing 100 U/mL penicillin and 100 μg/mL streptomycin and cells were maintained at 37°C in a humidified atmosphere of 5% CO2. Cells were cultured in hypoxic conditions, as previously described (Shen et al.,2007), in a tightly sealed hypoxic culture chamber (Changjin Science Technologies, Changsha, China). The cells were placed in the humidified chamber flushed with a gas mixture of 1% O2, 5% CO2 that was balanced with nitrogen (N2). The hypoxic chamber containing cell-culture dishes was placed in a culture incubator at 37°C or 4°C for the indicated time.
FACS Analysis and MTT Assay
Hypoxia-treated BEL-7402 cells were harvested, washed, and fixed in 75% ethanol in PBS at 4°C. The fixed cells were resuspended, and DNA was stained with 15 μg/mL propidium iodide (Sigma-Aldrich, St. Louis, MO). Cell cycle distribution assays were performed using flow cytometry for 10,000 events (FACS Calibur, BD, San Diego, CA). MTT assays were performed to determine the cell viability and proliferation. Briefly, BEL-7402 cells were plated at 1 × 104 cells/well in 96-well plates and incubated in normoxic or hypoxic conditions for 24 or 48 hr before the MTT assay. In some experiments, the BEL-7402 cells were infected with AdGFP or AdEgr-1 for 24 hr before seeding in the plates. The absorbance at 490 nm was determined using a Biokinetics plate reader (Bio-Tek Instruments, Winooski, VT).
Construction of Adenoviral Vectors and Adenovirus Infection
The replication-deficient adenoviruses encoding dominant-negative Egr-1 (pAd-dnEgr-1) were constructed according to the method of He et al. (He et al.,1998; Mesri et al.,2001). Briefly, the dominant-negative DNA fragment for Egr-1 containing 5′ HindIII and 3′ XbaI sites was inserted in pAdTrack downstream of the cytomegalovirus promoter to generate pAd-dnEgr-1. Each shuttle vector was linearized with PmeI, electroporated in Escherichia coli BJ5183, and colonies were selected in 50 μg/mL of kanamycin. Each pAd construct (4–10 μg) was digested with PacI, transfected in 293A cells by calcium phosphate, and cultures were monitored for expression of green fluorescent protein (GFP), by fluorescence microscopy. The cell pellets were suspended in 1 mL PBS, and after three cycles of freezing and thawing, 1 mL of viral lysate supernatant was used to infect 3 × 106 to 5 × 106 293A cells.
Cells were incubated with adenovirus in a small volume of serum-free medium at 37°C. After adsorption for 2 hr, fresh complete growth medium was added, and cells were placed in the incubator for additional time as indicated for the following experiments.
Immunofluorescence and Microscopy
Cultured cells were washed with PBS, fixed with 4% paraformaldehyde in PBS for 30 min, and permeabilized with 0.2% Triton X-100 in PBS for 10 min. The cells were incubated with a primary antibody against α-tubulin (sc-53646, 1:100 dilution, Santa Cruz, CA) or Egr-1 (sc-189, 1:30, dilution, Santa Cruz, CA) overnight at 4°C and then incubated with anti-mouse FITC-conjugated or rhodamine-conjugated secondary antibodies (Calbiochem, La Jolla, CA) for 2 hr at 37°C in the dark. After extensive washing, the cover slips were then mounted on glass slides and the fluorescent images were captured with a cooled charge-coupled device camera (Dignostic Instruments, MI) and processed using SPOT software (Dignostic Instruments).
For analyses of Egr-1 expression in BEL-7402 cells cultured under normoxic and hypoxic conditions, cells were extracted using RIPA buffer (50 mmol/L Tris-HCl, pH 7.2, 150 mmol/L NaCl, 0.1% SDS, 1% NP 40, 1 mmol/L Na3VO4, and 5 mmol/L EDTA) and protease inhibitor cocktail (Roche Diagnostics) for 30 min on ice. After determination of the protein concentration, 30 μg of total protein was resolved on SDS-PAGE gels and transferred to PVDF membranes (Roche, Mannheim, Germany). The membranes were probed with the appropriate antibodies (Egr-1, 1:100 dilution; α-tubulin, 1:400 dilution; Santa Cruz, CA), and immunoreactive protein complexes were detected by enhanced chemifluorescence (Amersham Pharmacia Biotech, Buckinghamshire, UK).
Egr-1 Knockdown and Inhibition
Egr-1 was knocked down with small interfering RNA (siRNA). The expression plasmid targeted to Egr-1 mRNA was purchased from Invitrogen (Carlsbad, CA). The siEgr-1 plasmid was transfected into BEL-7402 cells using Lipofectamine™ 2000 (Invitrogen, Carlsbad, CA) according to the manufacturer's protocol. An adenovirus-carrying dominant-negative Egr-1 (Ad-dnE) gene was used to suppress the transcriptional factor function of Egr-1 in BEL7402 cells. Cells were incubated with adenovirus-carrying GFP or adenovirus-carrying dominant-negative Egr-1 genes in a small volume of serum-free medium at 37°C. After adsorption for 2 hr, fresh complete growth medium was added and cells were placed in the incubator for the indicated times for subsequent experiments.
All data are expressed as means ± standard deviation. Statistical significance was determined by Student t test and values of P < 0.05 were considered significant. Statistical comparisons between different groups using analysis of variance (ANOVA) were performed with SPSS 8.0 software for Windows, and P values less than 0.05 were considered statistically significant.
Hypoxia Increased Egr-1 Expression and BEL-7402 Cell Viability
Hypoxia, a common feature of tumor and ischemic tissues, fundamentally and physiologically alters the activity of important intracellular pathways. Egr-1 is a hypoxia-sensitive gene (Murphy,2004), and its expression was elevated in BEL-7402 cells after exposure to hypoxia (Fig. 1A). The MTT assay showed that, compared with normoxia, the activity of BEL-7402 cells (Fig. 1B) was significantly increased by exposure to hypoxia for 12 hr. FACS analysis confirmed that BEL-7402 cells were enriched in the S stage (Fig. 1C) after exposure to hypoxia for 24 hr, which suggests that hypoxia enhanced cell proliferation.
Egr-1 Is Colocalized With Microtubules and Hypoxia Could Protect Tumor Cells From Microtubule Disassembly
Next, we determined the localization of Egr-1 after exposure to hypoxia. Egr-1 was found to translocate into the nucleus during the interphase in response to hypoxia (Fig. 2Aa). Meanwhile, during mitosis, from the prophase to the telophase, Egr-1 was localized on the spindle (Fig. 2Ab–e). The colocalization of Egr-1 with microtubules suggests that Egr-1 might regulate the function of the microtubules. When we depolymerized the microtubules with exposure to low temperature (4°C) (Fig. 2B) and nocodazole (Fig. 2C) treatment, hypoxia had a protective effect on the microtubule network. More microtubules were intact after exposure to low temperature during hypoxia than those kept at normal temperatures (Fig. 2Ba–d). Similarly, the microtubule network was kept intact after treatment with nocodazole during hypoxia (Fig. 2Ca–d). These findings suggest that hypoxia stabilizes the microtubule network of tumor cells.
Egr-1 Is Essential for Microtubule Protection Function of Hypoxia
To determine whether Egr-1 is involved in the protection of microtubules via hypoxia, siEgr-1 plasmid was transfected into BEL-7402 cells. The plasmid of siEgr-1 strongly downregulated hypoxia-induced Egr-1 expression (data not shown). When siEgr-1-transfected cells were exposed to low temperatures during hypoxia, the protective role of hypoxia on microtubules was lost and the microtubules were more sensitive to cold treatment (Fig. 3A). Because Egr-1 is a transcription factor, we next checked whether the Egr-1-mediated protection of microtubules was dependent on its transcriptional activity. Figure 3B shows that the dominant-negative Egr-1 (Ad-dnEgr-1) could not overcome the protective effects of hypoxia. This indicates that the protective effect of hypoxia on microtubule assembly was not dependent on its transcriptional activity.
Knockdown of Egr-1 Increased the Sensitivity of BEL-7402 Cells to Chemotherapeutics During Hypoxia
Because hypoxia normally promotes tumor cell resistance to chemotherapy, we next tried to determine whether knockdown of Egr-1 could increase the sensitivity of tumor cells to chemotherapy during hypoxia. The siRNA/Egr-1-transfected BEL-7402 cells were treated with vinblastine, a widely used chemotherapeutic that targets microtubules. As shown in Fig. 4, vinblastine successfully destroyed the microtubule network under normoxia but had no effect on microtubules under hypoxia. However, when Egr-1 was knocked down by siRNA, hypoxia could not protect microtubules from vinblastine-mediated disassembly (Fig. 4B). The MTT assay also showed that vinblastine decreased the activity of BEL-7402 cells exposed to hypoxia when Egr-1 was knocked down by RNAi. Taken together, our results suggest that knockdown of Egr-1 offers a novel approach that increases the sensitivity of tumor cells to chemotherapeutics that target microtubules.
Hypoxia is considered to be a potential barrier for treating tumors because it induces the development of cell clones with an aggressive and treatment-resistant phenotype, which leads to rapid progression and poor prognosis (Kizaka-Kondoh et al.,2003; Vaupel,2008). It has been shown that hypoxia can decrease the therapeutic efficacy of radiation treatment, surgery, and some forms of chemotherapy. These effects, in turn, can lead to increased invasiveness and metastatic potential, loss of apoptosis and chaotic angiogenesis, and further enhance treatment resistance of the tumor (Harrison and Blackwell,2004). Hypoxia deprives the cells of oxygen, which is essential for the cytotoxic actions of therapeutics, and contributes to the resistance of cancer cells to radiation therapy or chemotherapy. Hypoxia can indirectly contribute to radiation therapy and chemotherapy resistance by enhancing proteomic and genomic changes that ultimately result in malignant progression (Teicher,1995; Harrison et al.,2002). Although many successful approaches have been developed to overcome this mechanism of resistance, many of these approaches are associated with short- or long-term side effects. In this study, we found that hypoxia not only increased the viability and proliferation of human hepatocellular carcinoma cell BEL-7402 (Fig. 1B,C) but also improved the drug resistance of BEL-7402 cell by the protecting effect of Egr-1 to microtubules (Figs. 3, 4).
Microtubules form a mitotic spindle during cell division, which is the key machinery that drives the alignment of replicated chromosomes to the equatorial plane and mediates the subsequent segregation of chromosomes to the two daughter cells (Zhou and Giannakakou,2005). The progression of mitosis is arrested until the spindle has been correctly assembled during the metaphase through a regulatory system called the spindle checkpoint system (Sampath and Plunkett,2001). However, cancer cells often harbor defective cell cycle checkpoints allowing for uncontrolled cell proliferation, even when cell division does not occur properly. Therefore, effective cancer treatment can be achieved by drugs that target certain processes or proteins impinging on the cell cycle machinery. The critical roles of microtubules in cell division mean that they are a suitable target for the development of chemotherapeutic drugs against rapidly dividing cancer cells. Chemical compounds such as vinca alkaloids and taxanes that interfere with microtubules are used as powerful chemotherapeutic agents for the treatment of cancer (Sampath and Plunkett,2001; Harrison et al.,2002). However, because of the existence of hypoxia in solid tumors, the therapeutic efficacy of these drugs is largely decreased (see Figs. 2C, 4A) This has been proved in vitro and in vivo in a variety of tumors (Teicher,1995; Harrison et al.,2002). Therefore, there is a continued need for developing reasonable and effective strategies that improve regional tumor control and decrease their survival after chemotherapy.
The results presented here show that Egr-1 is rapidly upregulated during hypoxic conditions. Although Egr-1 translocates into nucleus in response to hypoxia, the Egr-1 remaining in the cytosol was found to be colocalized with microtubules during the interphase (Fig. 2A). This phenomenon was more obvious during mitosis when Egr-1 was localized to the spindle-related apparatus. Further studies showed that hypoxia was able to protect the microtubule networks within the tumor cells and increased the resistance of the tumor cells to antitumor drugs such as nocodazole and vinblastine. Meanwhile, knockdown of Egr-1 with siRNA reduced the protective effects of hypoxia on microtubules. Considering the above results, we speculate that the protective effects of hypoxia may be attributable to the distribution of Egr-1, which stabilized the microtubules during hypoxia.
Of particular interest, inhibition of Egr-1 transcription with dominant-negative Egr-1 did not reduce the protective effects of hypoxia. We still do not know how Egr-1 protects microtubules as a microtubule-associated protein in hypoxic conditions. The detailed mechanism by which Egr-1 is upregulated and protects microtubules during hypoxia still needs to be elucidated. Nevertheless, our results reported here suggest a novel mechanism by which hypoxia is involved in cancer progression and why some solid tumors are not sensitive to antitumor drugs that target microtubules. This newly described function of Egr-1 may provide valuable insight into cancer therapy.