2-Deoxy-D-glucose Mediates Dihydrodiol Dehydrogenases Over-expression and Cisplatin Resistance in Human Cervical Cancer Cells

Dihydrodiol dehydrogenases (DDHs) belong to a superfamily of cytosolic NADP (H)-dependent oxidoreductases. The over-expression of DDHs induces cisplatin resistance. 2-Deoxy-D-glucose (2-DG), a glucose analogue, is currently being used as an anticancer reagent. In this study we investigated the effect of 2-DG on DDHs expression and cisplatin resistance in human cervical cancer 2008 and C13 cells. We employed RT-PCR to detect mRNA level of DDH1, DDH2, and DDH3 and western blotting for protein expression. The cisplatin resistance was investigated with MTT and colony formation assays. Apoptosis/necrosis and mitochondrial membrane potential analysis were evaluated with flow cytometry. The intracellular ROS regulation was evaluated with H2DCFDA probe. We used 2-DG resistant cells to demonstrate the effect of 2-DG on DDHs expression and cisplatin resistance. 2-DG significantly up-regulated the mRNA level and protein expression of DDH1, DDH2, and DDH3, which consequently increased cisplatin resistance as confirmed by MTT and colony formation assays. In the 2DG-resistant cells, the apoptosis/necrosis percentage and intracellular ROS were significantly decreased. 2-DG itself could activate JNK. When treating the cells combined with cisplatin, 2-DG attenuated cisplatin-mediated JNK phosphorylation. 2DG down-regulated wild-type p53 protein expression at lower 2-DG concentrations (1/2 IC50 and IC50) at 24-hour. Activated JNK attenuation and decreased p53 expression by 2-DG implied the underlying resistance mechanism. Our study highlighted that 2-DG, as an anticancer reagent currently, could be a two-side sword that also significantly inhibited apoptosis by upregulating DDHs expression and consequently increased cisplatin resistance in the human cervical cancer cells we used.


Introduction
2-DG is a synthetic analogue of glucose. It can go into the glucose metabolic pathway and competitively inhibit hexokinase, which will affect intracellular glucose metabolism and ATP production. In recent decade, singly or in combination with other cytotoxic chemotherapeutic drugs, 2-DG has been widely used as an anticancer reagent in vitro or in some phase I/II clinical trials [1][2][3]. Even though the preliminary results from clinical trials were kind of discouraging, many researches are still going on to continuously report that 2-DG can significantly induce apoptosis, increase chemosensitivity, and even reverse drug resistance in different solid tumor and leukemia/lymphoma in vitro [5][6][7][8][9][10]. On the other hand, up to now only one report found that 2-DG was associated with P-glycoprotein (P-gp) over-expression in hepatoma cells, which might imply the unknown side of 2-DG in conferring the phenotype of multiple drug resistance (MDR) in cancer cells [4]. P-gp was known to be mainly associated with plant alkaloids resistance but not to cisplatin. To our knowledge, only one study reported that cisplatin could induce P-gp expression in animal cells but this never got confirmed in human cells [11]. As yet it is not clear whether 2-DG can induce cisplatin resistance. Cisplatin and carboplatin are amongst the most widely used drugs for the treatment of patients with solid tumors and are curative for most patients with germ cell tumors. They are currently widely used in standard chemotherapy protocols for the treatment of patients with ovarian, bladder, cervical, head and neck and lung cancers. However, the intrinsic and/or acquired resistance to these platinum compounds has increasingly become a main problem for successful management of these patients. Our previous studies demonstrated that the cytosolic enzyme DDHs were implicated in producing resistance to cisplatin [12][13][14]. Transfections of DDH cDNA sequences (primarily the DDH-1 isoform) into a whole series of human cancer cell lines derived from different primary tumors (ovary, cervix, germ cell, and lung) similarly produced cisplatin resistance [15] and its down-regulation with siRNAs correspondingly sensitized resistant cancer cells to cisplatin [13]. A small clinical study with human cervical cancer specimens employing immunohistochemistry showed that DDH expression may be associated with chemotherapy resistance [16]. And over-expression of DDH was even reported as a prognostic marker in non-small cell lung cancer [17].
In this study we investigated the effect of 2-DG on the expression of DDHs and cisplatin resistance. Our results demonstrated that 2-DG could mediate over-expression of DDHs, attenuation of both intracellular ROS and cisplatininduced JNK activation, down-regulation of nuclear p53, and over-expression of many other drug resistance related gene candidates, for example GRP78, HO-1, etc. To our knowledge this is the first study that 2-DG could induce cisplatin resistance, at least for sure in these human cervical cancer cells we used. Therefore, not only P-gp for MDR in hepatoma cells [4], 2-DG can also induce many other genes, for example DDHs, to mediate cisplatin resistance.

MTT Assay and Colony Formation Assay (CFA)
Cell cytotoxicity assays [18] were performed as described previously. Briefly, for MTT assay, 4000 Cells were seeded into 96-well plates in triplicates with different concentrations (0, 0.5, 1, 2, 5, 10, 20, 50µM) of cisplatin and incubated for 72 hours. 10 µl of MTT (5mg/ml) was added into each well and incubated for 5 hours at the end of the 72-hour incubation. The plates were read at 590 nm with a 96-well plate reader. For CFA, 200 Cells were seeded into 6-well plates and incubated for 24 hours. After treated with different concentrations (0, 0.1, 0.2, 0.5, 1, 1.5, 2, 5, 10 µM) of cisplatin for 4 hours, cells were washed twice and added fresh medium and followed by twoweek incubation to allow colony growth. At the end of incubation period, cells were fixed and stained with 0.5% methylene blue in 50% ethanol for 40 minutes at room temperature. Thereafter, the plates were gently washed with water and allowed to air-dry. Visible colonies (containing 50 or more cells each) were counted to determine the percent colony formation for drug treatment. IC50 values were calculated and expressed as the mean ± SD (standard deviation) from at least three separate triplicate experiments.

Apoptosis/Necrosis and Mitochondrial Membrane Potential Analysis
Apoptosis/necrosis was determined as described in the Annexin V-FITC apoptosis detection Kit I. Cells (1x 10 5 ) were pre-incubated in 500µl of complete RPMI-1640 medium for 24 hours in a 24-well plate. Cisplatin was added and incubated for another 24 hours. Then the cells were trypsinized and harvested and washed with HBSS buffer. The stained cells with Annexin and PI probes were analyzed on BD FACSCalibur with excitation wavelength 488nm: emission wavelength 525nm for annexin V and 575nm for PI.
To evaluate mitochondrial membrane potential depolarization occurring during cellular apoptosis, cells (1x 10 5 ) were pre-incubated in 500µl of complete RPMI-1640 medium for 24 hours in a 24-well plate. Cisplatin was added at different concentration (0, 20, 50, 100 µM) and incubated for 24 hours and JC-1 (final concentration at 10µM) added 30 min prior to the end of the incubation. The medium was aspirated, the cells trypsinized (0.05% trypsin-EDTA), centrifuged and the pellets washed and resuspended in HBSS buffer and analyzed with a BD FACSCalibur (excitation wavelength 525nm: emission wavelength 575nm).

Intracellular ROS Change by 2-DG-induced DDHs
For estimation of intracellular ROS levels, a cell membrane permeable and oxidant-sensitive fluorescent dye H 2 DCFDA was employed [13]. H 2 DCFDA, by itself, is non-fluorescent, but once inside the cells is hydrolyzed to H 2 DCF by nonspecific esterases. The H 2 DCFDA is a ROS sensitive intermediate whose degradation by intracellular hydrogen peroxides results in generation of fluorescent DCF [29]. Briefly, 4 x 10 4 cells were seeded in 96-well plates for 24-h incubation. Thereafter, the growth medium was aspirated and 100 µl of warm (37°C water bath-preincubated) HBSS buffer (pH 7.4) and H 2 DCFDA was added to each well (final concentration at 25 µM) and incubated further at 37°C for 30 min. The cells were then washed with warm HBSS twice and the generation of ROS (measured as fluorescence intensity) was measured on a multi-well fluorescence microplate reader (TECAN GENios, Durham, NC) with an excitation wavelength of 485 nm and an emission wavelength of 535 nm. The data were expressed as fluorescent intensity per 4 x 10 4 cells [13].

Semi-Quantitative RT-PCR Analysis
Total RNA was isolated from the cells after twice wash with cold PBS (pH 7.4). RNAzol (Tel-test, Friendswood, TX) was added followed by chloroform extraction, isopropanol precipitation, and a 75% (v/v) ethanol-DEPC wash. The reverse transcription reaction consisted of 1µg of RNA, 4 units of Omniscript RT, 1µM oligo-dT primer, 0.5mM dNTP, 10 units of RNase inhibitor, and 1 x RT buffer. Reverse transcription was performed at 37°C for 1h and inactivated at 93°C for 3 min. The cDNA was then amplified by PCR using gene specific primer pairs. Each PCR consisted of 1 x PCR buffer, 1.5mM MgCl2, 200µM dNTP, 2.5 units of Taq Polymerase and 0.2mM genespecific forward (F) and reverse (R) primers. The PCR conditions were as follows: an initial denaturation at 94°C for 15s, 55°C for 30s, 72°C for 30s for the number of cycles optimized for each primer to ensure that the product intensity fell within the linear phase of amplification, and then a final elongation step was performed for 10min at 72°C. RT-PCR amplification of GAPDH was used as internal control to verify that equal amounts of RNA were used from each cell line. All primer sequences are as follows [13]:

Western Blotting Analysis
Cells at a density of 1 x 10 6 /ml were incubated under normal growth conditions and then washed with chilled PBS (3x) and a whole cell lysate prepared from each of the cell lines by scraping the cells into a buffer containing 20 mM Tris-HC1 pH 7.5, 150 mM NaCl, 1mM EDTA, 1mM EGTA, 1% (v/v) Triton X-100, 0.5% (v/v) Nonidet P40, 2.5mM sodium pyrophosphate, 1mM sodium orthovanadate, 50 mM sodium fluoride and 1x protease inhibitor cocktail and incubated on ice for 30min. The lysate was then centrifuged at 13,000g for 10min and the supernatant was transferred to a fresh tube and stored at -80°C until use. Nuclear protein extraction for p53 followed the manual instructions of Nuclear Extraction Kit (Millipore, MA). Proteins (10 or 30µg) were loaded and separated by SDS-PAGE and were transferred to a PVDF membrane. The different antibodies were applied (at concentrations described by the manufacturer) to the PVDF membrane and the bands identified by enhanced chemiluminescence reagents (Pierce Biochemicals, Rockford, IL). The antibodies utilized were goat polyclonal GRP78, rabbit polyclonal JNK, phospho-JNK (Santa Cruz Biotechnology-Santa Cruz, CA), DDH-1 and DDH-2 mouse polyclonal antibodies (Abnova Corp. -Walnut, CA), mouse monoclonal p53 (St. Louis, MO) and mouse monoclonal actin (Ab-1) (Calbiochem-San Diego, CA) and a DDH-3 mouse monoclonal antibody (Abcam -Cambridge, MA). Anti-NF-YA polyclonal antibody and the secondary antibodies were anti-goat, anti-rabbit, and anti-mouse (Pierce Biotechnology-Rockford, IL). Densitometry analysis was done with Image J software.

Statistical Analysis
The linear regression analysis and paired t-test were performed using the SigmaStat Statistical Analysis System, Version 1.01. P<0.05 was considered significant.

Growth Inhibition by 2-DG Treatment
Similar to other study [1], our results also found 2-DG inhibited the growth of these sensitive and resistant human cervical cancer cells. When the maximal physiological concentration (8.88mM in RPMI-1640 medium [19]) of 2-DG was administered, the growth inhibition was even more significant than that by cisplatin at 2µM in sensitive 2008 cells or at 20µM in resistant C13 cells (Figure 1a, 1b). When 2mM and 4mM of 2-DG were used in combination with cisplatin, IC50s of cisplatin by MTT assay would be decreased by 2-3 times in 2008 cells or by 6-16 times in C13 cells (Table 1). These findings suggested that 2-DG could significantly inhibit the growth of these cells and increase their sensitivity to cisplatin. Interestingly, the resistant C13 cells were even more significantly inhibited at the same concentration (2mM or 4mM) when compared with the parental sensitive 2008 cells. This phenomenon was also observed in other study [29] and was believed to be associated with the higher proliferation rate in the resistant cells than in sensitive cells. IC50 of 2-DG in 2008 by MTT assay was 2.5mM but in C13 3.3mM, with only 0.8mM difference between these two cell lines.

Up-regulation of mRNA and Protein Expression of DDHs by 2-DG in both Time-dependent and Dose-dependent Modes
Our previous studies have found that DDH1, DDH2, and DDH3 are associated with cisplatin resistance in many cancer cell lines with different tissue origins [12,14]. Over-expression of DDH1-3 by full-length cDNA transfections increased cisplatin resistance but knockdown of DDHs with siRNAs decreased cisplatin resistance [12,13]. Here we detected the effect of 2-DG on mRNA and protein expression of DDH1-3. We used IC50s of 2-DG (2.5mM for 2008; 3.3mM for C13) to treat cells for 0, 30min, 2h, 6h, 8h, 24h, 48h, and 72h. We found that mRNA levels of DDH1-3 were up-regulated by a time-dependent mode in both 2008 and C13 cells (Figure 2a,  2b). Also, we treated cells with 0, 1/8 of, 1/4 of, 1/2 of IC50, IC50, and 8.88mM (maximal physiological concentration) of 2-DG for 6 hours. Our results showed the similar finding that mRNA levels of DDH1-3 were also increased by a dosedependent mode in both cells (Figure 2c, 2d). To compare with the effect of cisplatin on DDHs mRNA, we treated 2008 and C13 cells for 6 hours with different concentrations of cisplatin. Surprisingly, we found cisplatin could only slightly upregulate DDH1mRNA in 2008 at a relatively higher concentration of cisplatin (20µM in 2008 cells) but no mRNA change was found in C13 cells even at as high as 100µM of cisplatin (Figure 2e, 2f). These findings implied that 2-DG was more easily than cisplatin to increase the expression of DDHs. Likewise, we treated cells in a similar way to check DDH1-3 expression by western blotting and we got the very similar results showing DDH1-3 expression was consistently upregulated by 2-DG in a time-or a dose-dependent mode in these cells (Figure 3a, b, c, d).   In view of the possibility that the up-regulation of DDH1-3 might be a reversible short-term event, we continuously treated 2008 and C13 cells at 2.5mM and 3.3mM of 2-DG respectively for 6 month until they could stably grow in RPMI-1640 medium with 2-DG. We named these cells as 2008/2.5mM and C13/3.3mM. Their growth doubling cycles were similar to those of their parental cells. We checked mRNA levels (Figure 4a, 4b) and protein expression of DDH1-3 (Figure 4c, 4d), and found the up-regulation of DDH1-3 was stably confirmed in these cells.

2-DG-mediated Over-expression of DDHs Increases Cisplatin Resistance
To evaluate the effect of 2-DG on cisplatin sensitivity, we investigated the cytotoxicity of 2008, 2008/2.5mM, C13, and C13/3.3mM cells to cisplatin by MTT and colony formation assays. These assays gave us very similar results that 2008/2.5mM cells showed increased cisplatin resistance by 2 times and C13/3.3mM cells by 1.5 times when comparing with their parental cells. Based on our previous study results that over-expression of DDHs is consistently associated with increased cisplatin resistance [12,13,15], our current results clearly suggested that 2-DG up-regulated DDH1-3 expression, which consequently caused an increase of cisplatin resistance based on the significant increasing change of IC50s of cisplatin in these 2-DG-resistant cells by both MTT and CFA assays (Table 2a, 2b).

Decreased Intracellular ROS Level in 2-DG-resistant 2008 and C13 Cells
Functional cytosolic DDHs attenuate intracellular ROS level [13]. To see whether 2-DG-resistant 2008/2.5mM and C13/3.3mM cells express functional DDHs attenuating intracellular ROS, we checked their intracellular ROS level with H 2 DCFDA probe. The results suggested that with the increasing DDHs expression there were corresponding lower intracellular ROS levels in 2008/2.5mM and C13/3.3mM cells when compared with their parental 2008 and C13 cells [ Table 4]. This change of intracellular ROS level was completely compatible with our previous finding that intracellluar ROS level was inversely associated with cytosolic DDHs level.

Decreased Apoptosis/Necrosis by 2-DG-mediated DDHs Over-expression
To further check 2-DG-induced drug resistance and cellular apoptosis, we detected the effect of 2-DG on apoptosis/necrosis in these cells. Firstly we checked the effect of 2-DG alone by treating the cells with 2-DG at 0, 5mM, and 8.88mM for 24 hours. We found that 2-DG treatment alone could not induce significant change of apoptosis/necrosis (Table 3a). Simons et al [19] reported that ROS contribute to the toxicity induced by 2-DG alone or in combination with cisplatin. N-acetyl-cysteine (NAC) can inhibit any effect caused by 2-DG in their study. Considering this as a possible reason, we used NAC (5mM) to evaluate the effect on apoptosis. Surprisingly, our findings showed that NAC did not inhibit but dramatically induced the apoptosis/necrosis in both cells (Table 3a). This is compatible with some reports that NAC could increase apoptosis by direct actions on apoptotic pathways [30] or by endothelial nitric oxide synthase-mediated production of nitric oxide [31]. Therefore, our finding does not support the hypothesis that ROS generation upon 2-DG treatment is the underlying factor for inducement of apoptosis/necrosis. Cell type dependence might be the reason for the above discrepancy with other authors.
In addition to the effect of 2-DG alone on apoptosis/necrosis above, we also treated 2008, 2008/2.5mM, C13, and C13/3.3mM cells with cisplatin at 50µM for 24 hours. Our results suggested that the apoptosis/necrosis rate was significantly decreased in 2008/2.5mM cells (10.86%) when compared with that in 2008 cells (23.36%) (P<0.05). No significant apoptosis/necrosis decrease was observed in C13/3.3mM cells when compared with C13 cells at 50 µM cisplatin treatment (Table 3b). However, when we treated with higher concentration of cisplatin (400 µM), the apoptosis/necrosis percentage in C13/3.3mM cells showed 2 times less than that in its parental C13 cells (7.42% vs 15.3%, P<0.05) (Table 3b).
To double check the effect above, we employed molecular probe JC-1 to detect cellular mitochondrial membrane potentials. When mitochondrial membrane depolarization was measured following 24-hour cisplatin treatment, a good correlation with apoptosis/necrosis was observed. The 2008/2.5mM and C13/3.3mM cell lines showed lower degree of depolarization on cisplatin treatment when compared with those of 2008 and C13 cells (35.1% in 2008 vs 16.76% in 2008/2.5mM at 50µM of cisplatin; 73.82% in C13 vs 56.32% in C13/3.3mM at 150µM of cisplatin; both P<0.05).

Phosphorylation of JNK by 2-DG and Its Relationship with DDHs Expression
To ask the effect of 2-DG on JNK phosphorylation, we treated 2008 and C13 cells with IC50 (2.5 mM and 3.3 mM respectively) and the maximal physiological concentrations (8.88mM). Results suggested that 2-DG itself could activate JNK phosphorylation (Figure 5a). Since 2-DG up-regulates DDHs and activates JNK, we asked if JNK phosphorylation was an early event upon 2-DG treatment before the upregulation of DDHs, which might also be one of JNK downstream substrates including ATF-2, c-Jun, Bcl-2, Bcl-XL, p21, p53, etc [20]. We treated 2008 cells with 10µM of JNK inhibitor SP600125 (Calbiochem, CA). We speculated if JNK inhibitor could inhibit DDH up-regulation, we would know that the up-regulation of DDHs induced by 2-DG is initially mediated by earlier JNK activation. Our results found that JNK inhibitor itself or its solvent DMSO had no any effect on DDH1 expression. However, when treating 2008 cells with 2-DG together, JNK inhibitor could synergistically up-regulate DDH1 expression with 2-DG ( Figure 5b). More interestingly, both cisplatin and 2-DG could activate JNK separately, however, when treating cells with 2-DG in combination with cisplatin, 2-DG would attenuate cisplatin-mediated JNK phosphorylation in 2008 cells (Figure 5c). We previously found that JNK phosphorylation was associated with increasing apoptosis in 2008 cells [14], so both the inhibition of cisplatin-induced phosphorylation of JNK and the up-regulation of DDHs by 2-DG treatment would definitely contribute to inhibiting apoptosis and increasing cisplatin resistance in these cells.

Down-regulation of Wild-type p53 Protein by 2-DG Treatment
Both 2008 and C13 cells express wild-type p53 and cisplatin treatment can significantly up-regulate p53 protein expression in both cell lines [21]. Since wild-type p53 is widely known to play an important role in apoptosis regulation, we asked if 2-DG singly or in combination with cisplatin also would increase p53 expression. We initially treated 2008 cells with 10µM of cisplatin and 2.5mM of 2-DG for 24 hours. As expected, cisplatin apparently increased p53 expression. However cells. However, when the cells were treated for 48 hours, p53 expression increased at lower concentrations (½ IC50 and IC50) but decreased at higher 8.88mM (Figure 6b). p53 is mostly regarded as a nuclear substrate of phosphorylated JNK relocated from cytoplasm after activation [20]. We asked if p53 expression change by 2-DG in our cells was mediated only through JNK pathway. If this was the case in our cells, 2-DG-activated JNK phosphorylation should have increased p53 expression, not the opposite. Furthermore, if JNK inhibitor SP600125 (10µM) was used to treat the cells with 2-DG, p53 should have been decreased even more plus the effect from 2-DG. To investigate this hypothesis, we treated 2008 cells with JNK inhibitor SP600125 (10µM) with and without 2-DG (2.5mM) for 24 hours. Our results showed that p53 expression had no any change with 10 µM of SP600125 itself but significantly increased when SP600125 was used with 2-DG together (Figure 6c). These findings suggested that the downregulation of p53 expression by 2-DG might be independent of JNK pathway in our cells. Wild-type p53 is widely known to play an important role in regulating apoptosis. Even though it could be an early event (24 hour only), the decreasing wild-type p53 expression by 2-DG could be associated with the decreased apoptosis/necrosis and cisplatin resistance in our cells.

Discussion
Cisplatin and carboplatin are widely used for treatment of many solid tumors including cervical cancer. However, the therapeutic outcome is often disappointing due to relapse to resistant disease. During the recent decade, 2-DG has been found to be able to inhibit the growth of tumor cells in vitro [1]. Several clinical trials were also conducted to evaluate its effectiveness for cancer treatment alone or in combination with other chemotherapeutic drugs but finally did not get promising results [2,3]. Here we showed for the first time that 2-DG could significantly up-regulate the expression of DDHs and increase cisplatin resistance in human cervical cancer cells.
The JNK/p38 pathway was previously demonstrated to be involved in cisplatin cytotoxicity in 2008 and C13 cells [22]. A relationship between cisplatin treatment and phosphorylation of JNK and p38 was shown with phosphorylation being greater in the sensitive cell lines as opposed to the resistant ones following 24-hour cisplatin treatment. The phosphorylation of JNK/p38 consequently activates downstream events inducing apoptosis/necrosis and cytosensitivity to cisplatin [22]. To try to elucidate the mechanism of 2-DG-induced DDH over-expression and the decreased apoptosis/necrosis, we detected the JNK activation. Similar to cisplatin, 2-DG could also activate JNK alike. However, when cells were treated with 2-DG and cisplatin together, 2-DG significantly attenuated cisplatininduced JNK phosphorylation, which implied the mechanism for cisplatin resistance triggered by 2-DG. The role of JNK in apoptosis may stem from its ability to regulate the function of p53. In non-stressed state, it is found that JNK associates with p53 and targets it for ubiquitination and degradation [23]. When cells are exposed to stress or DNA damage, the phosphorylated JNK activates p53 and finally induces apoptosis/necrosis. These 2008 and C13 cells express wildtype p53. It has been reported that higher level of wild-type p53 enhances the apoptotic death of prostate cancer cell lines PC-3 and DU-145 [24]. Our study found that 2-DG downregulated wild-type p53 expression at lower concentration (IC50) for 24-hour treatment, which could have subsequently explained the decreased apoptosis/necrosis in 2-DG-induced cisplatin resistance. The underlying mechanism of these cellular responses to 2-DG alone or combined with cisplatin is unknown. Traditionally, GRP78 up-regulation has been regarded as the hallmark of the endoplasmic reticulum (ER) stress and unfolded protein response (UPR) on 2-DG treatment [25], we also found GRP78 expression up-regulated in our cells in a dose-dependent mode by 2-DG in our cells (Figure 7a, 7b). In multiple tumor types including lung, bladder, stomach, breast, gastric, and epidermoid carcinoma, GRP78 overexpression has been found to confer resistance to a wide variety of chemotherapeutic agents, and knockdown of GRP78 sensitizes the tumor cells to drug treatment [26,27].
So besides of DDHs, GRP78 over-expression by 2-DG in our cells might also contribute to drug resistance. By RT-PCR our preliminary results did not found that cisplatin could induce any change of GRP78 mRNA (Figure 7c). Furthermore, we also found that 2-DG could cause overexpression of HO-1, thioredoxin1, and GSTpi in our cells (data not shown here). All of these proteins have been well known to be associated with cancer chemoresistance. So all in all our findings suggested that 2-DG could cause changes of a serial of other drug resistance related genes, which could also at least partially confer the increasing chemoresistance. We previously found that the transcription factor (NF-Y) preferentially bound to the inverted CCAAT box at -109ATTGG-105 of the DDH-1 gene. A two-fold increase with cisplatin treatment on transcription activity was found with 2008 cells [28]. Since 2-DG increases DDH1 expression and the CCAAT box has been more and more focused on as a general transcription regulation site [28], we would like to check whether 2-DG could regulate NF-YA and consequently affect its binding activity to DDH proximal promoter. Our preliminary result showed that 2-DG at 8.88mM significantly increased NF-YA protein expression in 2008 cells and slightly increased in C13 cells at IC50 (3.3 mM) concentration (Figure 7d). So the effect of 2-DG on DDHs gene transcription regulation is also under way in our laboratory.
Hence in summary, since the precise molecular mechanisms of 2-DG treatment are not clearly known, we would ask the question whether 2-DG is safe to be used clinically or whether the benefits from 2-DG combination are necessarily over the potential risks. Based on the findings in this study, we provided a cautionary note and highlighted 2-DG might significantly cause cisplatin resistance and inhibit apoptosis by inducing DDHs over-expression.