To ensure the success of systemic gene therapy, it is critical to enhance the tumor specificity and activity of the promoter. In the current study, we determined that topoisomerase IIα promoter is selectively activated in breast cancer cells. An element containing an inverted CCAAT box (ICB) was shown to be responsible for the breast cancer specificity. When the ICB-harboring topoisomerase IIα minimal promoter was linked with an enhancer sequence from the cytomegalovirus immediate early gene promoter (CMV promoter), this composite promoter, CT90, exhibited activity comparable to or higher than the CMV promoter in breast cancer cells in vitro and in vivo, yet expresses much lower activity in normal cell lines and normal organs than the CMV promoter. A CT90-driven construct expressing BikDD, a potent proapoptotic gene, was shown to selectively kill breast cancer cells in vitro, and to suppress mammary tumor development in an animal model of intravenously administrated, liposome-delivered gene therapy. Expression of BikDD was readily detectable in the tumors but not in the normal organs (such as heart) of CT90-BikDD-treated animals. The results indicate that liposomal CT90-BikDD is an effective systemic breast cancer-targeting gene therapy.
Breast cancer is the most common cancer and the second leading cause of cancer-related death for women in the United States.1 The 5-year survival rates for patients with local or regional diseases are about 80% or higher. For patients with metastasis, however, the survival rate drops to about 20%. Many patients experience disease recurrence even after adjuvant therapy such as hormone therapy, radiotherapy or chemotherapy. Therefore, the development of new therapies is a very important clinical issue in breast cancer. As micrometastasis occurs frequently in breast cancer, newly developed therapies must be delivered systemically.
Almost all types of cancers exhibit genomic instability,2, 3, 4 resulting in mutation, loss or gain of genes. Gene therapy suppresses tumors by transferring a gene into cancer cells to correct their genetic defects, block their abnormal signaling or induce their death. In recent years, advances in DNA delivery technology, especially cationic lipids, have made systemic delivery feasible for gene therapy. Cationic liposome is easy to produce and is relatively nontoxic. Its low immunogenicity allows its repeated administration to patients, increasing overall efficiency of gene delivery and therapeutic efficacy. Multiple preclinical studies and clinical trials have demonstrated the therapeutic efficacy and minimal side effects of liposomal gene therapy for cancers.5, 6, 7, 8
The efficacy of gene therapy depends on the promoter activity. As breast cancer frequently develops metastasis, the new therapy should be designed for systemic administration. Several studies have pointed out that, to ensure efficacy for systemic gene therapy, promoter activity in the targeted cells should be comparable to that of the promoter of cytomegalovirus immediate early gene (CMV promoter).10, 11, 12 In fact, the CMV promoter is the most widely used promoter in gene therapy settings,10, 13 especially those in clinical trials (according to The NIH/FDA Genetic Modification Clinical Research Information System (GeMCRIS; www4.od.nih.gov/oba/rac/GeMCRIS/GeMCRIS.htm) and The Journal of Gene Medicine Clinical Trial site (www.wiley.co.uk/genetherapy/clinical/)). However, the CMV promoter has ubiquitous activity; it drives high level of gene expression in both cancer and normal cells. When cationic liposome is used to deliver CMV promoter-driven reporter gene systemically, significant uptake of the DNA construct occurs in lung and heart, resulting in a high level of reporter activity in these organs.14, 15, 16, 17 Undesired gene expression in normal organs may cause side effects in cancer gene therapy.10, 11 Strategies to reduce such risk for liposomal gene therapy include the adoption of a promoter that induces a higher level of gene expression in tumor cells and a lower level in normal cells, thus decreasing undesired gene expression secondary to systemic treatment. For systemic treatment purposes, therefore, the promoter in gene therapy should possess (1) activity higher than or comparable to that of the CMV promoter in tumor cells and (2) a higher cancer specificity index (ratio of promoter activity in tumor to that in normal tissue (tumor:normal))18 than the CMV promoter. Several promoters have been used to drive breast cancer-targeting gene therapy in cell lines or animal models,13 such as the human alpha-lactalbumin promoter, ovine beta-lactoglobulin promoter,19 type II hexokinase promoter,20 HER2/neu promoter9 and hTERT promoter.21 However, the activities of these promoters are in general weaker than the CMV promoter. Identification of a promoter with these two properties improves the safety of systemic gene therapy protocols while achieving sufficient therapeutic efficacy.
In the current study, we identified a promoter with these two properties, and used it to develop a liposome-delivered breast cancer-targeting gene therapy for systemic administration. To accelerate the clinical use of this system, we adopted materials that have already been tested in clinical trials, including backbone DNA vector pUK215, 22 and extruded DOTAP:Chol liposome (Human Gene Transfer Protocol 0201-513, NIH Genetic Modification Clinical Research Information System).14, 15 As a first step, we determined that the topoisomerase IIα (topoIIα) promoter is specifically activated in breast cancer cells, and that its cis-element, the inverted CCAAT box (ICB),23, 24, 25 mediates this specific activity. To enhance its basal activity level, we linked the minimal topoIIα promoter harboring ICB1 with the enhancer sequence from the CMV promoter. We demonstrated that the activity of this composite promoter, CT90, is comparable to that of the CMV promoter in breast cancer cells, but much lower in normal cells in vitro and in vivo. The CT90 promoter was then used to drive a potent proapoptotic gene, BikDD,26 in the pUK21 vector (CT90-BikDD). We further showed that the intravenously injected DOTAP:Chol liposome–CT90-BikDD complex effectively suppressed tumor growth and extended animal lifespan in orthotopic breast cancer mouse models. These results demonstrated the tumor-targeting and therapeutic efficacy of the DOTAP:Chol–CT90-BikDD complex when administered systemically, which has important clinical implications. Moreover, as most of its materials have been tested in clinical trials, this cancer gene therapy regimen has great potential for transfer to clinical use.
Materials and methods
Human cancer cell lines (breast cancer MCF-7, MDA-MB-231, MDA-MB-468, MDA-MB-453, SK-BR-3 and T47D; ovarian cancer SKOV-3.ip-1; non-small-cell lung cancer A549; pancreatic cancer Capan-1) and normal human cell lines (breast epithelial cell line 184A1, normal fibroblast line WI-38 and nonmalignant liver cell line Chang liver) were all purchased from American Type Culture Collection (Manassas, VA), and cultured according to the vendor's instructions.
Liposome and DNA constructs
Extruded DOTAP:Chol liposome (20 mM) from the nonviral core facility at Baylor Collage of Medicine was used. It was prepared following the protocol described previously.15 The defined full-length −572/+90 sequence of the topoIIα promoter was amplified via PCR from genomic DNA of HeLa cells by using primers T2A-F (5′-IndexTermGGGGCGGGGTTGAGGCAGATG-3′) and T2A-R (5′-IndexTermCCATGGTGACGGTCGTGAAGGGGC-3′), and then ligated into the pCR2.1-TOPO TA cloning vector (Invitrogen, Carlsbad, CA). The −572/+90 sequence was released by digesting the TA vector at HindIII/XhoI and ligated into pGL3-basic (Promega, Madison, WI) at the corresponding sites to construct the topoIIα promoter-driven luciferase reporter topoIIα-pGL3. The deletion mutants of the topoIIα promoter were amplified by PCR from topoIIα-pGL3 and subcloned into the pGL3-basic vector to generate T2A354-pGL3, T2A182-pGL3, T2A142-pGL3, T2A90-pGL3 and T2A60-pGL3. To generate the ICB1-mutated topoIIα promoter, two fragments were PCR amplified by using two sets of primers: T2A-F and 5′-IndexTermCAGACCAGCCTTTCCCTGAC-3′, as well as 5′-IndexTermGTCAGGGAAAGGCTGGTCTG-3′ and T2A-R. The full-length mutant fragment, mICB1-T2A, was obtained in secondary PCR by using the mixture of the primary PCR products as template, and T2A-F and T2A-R as primers. The sequence of 214–645 bp in the pCDNA3.1 vector (Invitrogen), which contained the CMV enhancer sequence (−604/−173 in the CMV promoter),27 was PCR amplified with primers 5′-IndexTermTTGCTAGCTACGGGCCAGATATACGCGTT-3′ and 5′-IndexTermAACTCGAGTGAGTCAAACCGCTATCCACG-3′. The PCR product was digested with NheI/XhoI and subcloned at corresponding sites into topoIIα-pGL3, T2A182-pGL3 and T2A90-pGL3 to obtain CT572-pGL3, CT182-pGL3 and CT90-pGL3, respectively. The BikDD sequence was PCR amplified from the CMV-BikDD expression vector26 and subcloned into a HindIII/XbaI-digested CT90-pGL3 vector to obtain CT90-BikDD. For antitumor activity assay, CT90-BikDD and CMV-BikDD were PCR amplified and subcloned into pUK21 vector (GenBank accession no. AF223640). Expression vectors of cdk2 and dominant-negative cdk2 mutant (cdk2-dn) were kindly provided by Dr Sander van den Heuvel (Massachusetts General Hospital, Boston, MA). Cyclin A (CCNA) expression vector was a kind gift from Dr Brenda Schulman (St Jude Children Research Hospital, Memphis, TN). Cyclin E expression vector was a generous gift from Dr Khandan Keyomarsi of our institute.
In vitro promoter activity assay
The promoter activity in the cell lines was determined by transient transfection and reporter assay. Cell lines were cultured in six-well tissue clusters to 80% confluency. Tested plasmid (1.5 μg) and 0.075 μg of pRL-TK (Promega), which was used to normalize transfection efficiency, were co-transfected into cells by the liposome method described previously.6 Twenty-four hours after transfection, cells were harvested and cell lysates were prepared. The reporter activity in the cell lysates was determined by using the dual luciferase reporter assay system (Promega) according to the manufacturer's instructions. Three independent experiments were performed for each data point reported.
In vitro luciferase viability assay
The cell-killing effect of the gene therapy construct was determined by in vitro luciferase viability as described previously.26 Briefly, cell lines in six-well plates were co-transfected with gene therapy constructs (CT90-BikDD or CMV-BikDD) in different doses (0, 1.0 or 3.0 μg), and with CMV-luc (50 μg), the CMV promoter-driven luciferase reporter construct). At each dose, the pCDNA3.1 vector was replenished to keep the total amount of transfected DNA equal (3.05 μg). Twenty-four hours after transfection, cells were harvested and cell lysates were prepared. Luciferase activities were measured by using the single luciferase assay system (Promega) according to the manufacturer's instructions. The proportions of surviving cells were calculated by dividing the reporter reading from the experiment group (therapeutic construct 1.0 or 3.0 μg) by that from control group (no therapeutic construct) and calculating the percentage. Values are means±s.d. of three independent experiments.
Tissue distribution of the liposome-delivered reporter genes
The human breast cancer cells (MDA-MB-231 or MDA-MB-468) were trypsinized, harvested, and 106 cells were inoculated into the mammary fat pads of nude mice (nu/nu, 7–8 weeks old, 18–22 g; Charles River Laboratories, Wilmington, MA) to induce tumors as described previously.18 Five weeks later, the tumor-bearing mice were divided into two groups (five mice per group). CT90-pGL3 or CMV-pGL3 plasmid (50 μg) was mixed with DOTAP:Chol as described elsewhere15 for each mouse. The DNA–liposome complex was injected into each mouse through the tail vein. Forty-eight hours after injection, tumors and normal organs, including heart, lung, liver, spleen, kidney and muscle, were resected and immediately frozen on dry ice. The collected tissues were homogenized with 1 × passive lysis buffer (Promega) and subjected to centrifugation, and the supernatants were collected for measurement of luciferase activity and protein concentration. Protein concentrations were determined by using the Bio-Rad protein assay reagent (Bio-Rad, Fremont, CA). The luciferase activity per microgram of protein was used to compare gene expression in different tissues. The background levels of luciferase activity in tumors and organs were measured in tumor-bearing mice intravenously injected with the empty pGL3–liposome complex.
In vivo antitumor activity
The antitumor activity assay was performed following the principles of a previous study.18 The human breast cancer cells were inoculated into nude mice to induce tumors, as described in the above section. Mice were divided into two groups to receive injections of 15 μg of DNA–liposome complex once per week. In each group, mice were further divided into four subgroups of eight mice, and each subgroup received a different treatment (CT90-BikDD, CMV-BikDD, PGL3 vector or 5% dextrose). Tumor volume was measured on days 1 and 4 in each week. The tumor growth ratio, as described previously,18 was determined by comparing the size of the treated tumors at various times with the size of the original tumor before any treatment. The tumor volume (V) was calculated using the formula V=SSL/2, where S is the short length of the tumor in mm and L is the long length of the tumor in mm. The tumor volume ratio (R) was calculated as R=VN/V0, where VN is tumor volume at day N and V0 is tumor volume at day 0. The tumor volume ratio was introduced to normalize the individual difference of tested animals as described previously.18 The mice received a total of eight injections in the QW group and 12 injections in the BIW group. The tumor volume and survival data were analyzed with two-sided log-rank statistical tests and Kaplan–Meier analysis with log-rank test, respectively, using statistical analysis software SPSS (SPSS Inc., Chicago, IL).
In situ hybridization
Mice were prepared and inoculated as described previously. After the tumors grew to an average of 500 mm3, CT90-BikDD or CMV-BikDD was mixed with DOTAP:Chol and injected intravenously at a dosage of 50 μg of DNA per mouse in each group. Forty-eight hours after injection, the mice were killed. Tumors and several organs were resected, fixed in 10% buffered formalin for 48 h, and then processed and embedded in paraffin. The BikDD coding region sequence was obtained by digesting the CMV-BikDD vector at Hind III/Xba I and subcloned into the corresponding sites in the pSPT18 vector (Roche Diagnostics, Indianapolis, IN). The antisense and sense BikDD hybridization probes were synthesized by in vitro transcription using SP6 and T7 polymerase, respectively, in the presence of digoxygenin-UTP from the DIG RNA labeling kit (Roche Diagnostics, Indianapolis, IN) following the manufacturer's instructions. Hybridization was performed as described.28, 29 Signals were detected using the DIG nucleic acid detection kit (Roche Diagnostics, Indianapolis, IN), following the instruction manual.
TopoIIα promoter activity is specifically activated in breast cancer cells
As the first step in identifying a promoter with higher specificity for breast cancer cells, we searched for genes known to be overexpressed in breast tumor tissues. Although mechanisms of overexpression include gene amplification, transcriptional and translational upregulation, or increased stability of mRNA and protein, we reasoned that, for certain genes, overexpression might result from transcription upregulation, and the activities of their promoters could be much stronger in breast cancer cells than in other types of cells. With this notion in mind, we sought genes overexpressed in breast cancer cells from published data of gene expression profile30, 31, 32, 33 and serial analysis of gene expression (SAGE).34, 35 We selected those overexpressed genes commonly shown in these studies and identified four candidate breast cancer-specific genes: topoisomerase IIα (topoIIα), B-myb, ceruloplasmin and transferrin receptor. We obtained or cloned their promoters and examined their activities in normal and breast cancer cell lines with a reporter assay (Table 1). As an initial screening, we selected three breast cancer cell lines (MDA-MB-468, MDA-MB-453 and MDA-MB-231), a normal lung fibroblast cell line (WI-38) and a nonmalignant hepatocyte (Chang liver).36 We selected the lung fibroblast cell line as a representative normal cell line for comparison of promoter activity in this quick screening because lung is the major nonspecific targeted organ when a gene is delivered intravenously by cationic liposome.14, 15, 16, 17 On average, the topoIIα promoter possessed the highest level of activity in breast cancer cells and the lowest level of activity in normal or nonmalignant cell lines, suggesting that it has the highest degree of breast cancer specificity. The other promoters had either a lower level of activity in breast cancer cells (ceruloplasmin and transferrin receptor) or a higher level of activity in normal cells (B-myb). Therefore, we used the topoIIα promoter to develop a breast cancer-targeting gene therapy setting in the following experiments.
To test its specificity for breast cancer cells, we then examined the activity of a defined full-length (−572 to +90) topoIIα promoter25 in an expanded panel of cell lines by reporter assay. The results confirmed that, on average, topoIIα promoter possessed a higher level of activity in breast cancer cells and a lower level in other cell types (Figure 1a), including immortalized normal cell lines (184A1, WI38 and Chang liver) and other cancer cells (CAPAN-1, SKOV3-ip1 and A549), fulfilling one property of the promoter for systemic gene therapy. We then measured the ratio of topoIIα promoter activity to CMV promoter activity (Figure 1b, topoIIα/CMV). We observed that topoIIα/CMV ratios in most breast cancer cell lines are higher than those in other types of cell lines. It is interesting to note that the topoIIα/CMV ratio is also relatively high in SKOV-ip1 cell line, which is an ovarian cancer cell line harboring overexpression of HER2. The activity of topoIIα promoter in breast cancer cells was much weaker than that of the CMV promoter (<2.5%), however, and thus was expected to be insufficient for therapeutic efficacy, especially in the systemic treatment.
ICB1 can mediate activation of topoIIα promoter by Cdk2/cyclin A specifically in breast cancer cells
One straightforward approach to enhance the basal activity level of topoIIα promoter in breast cancer cells is to link a strong enhancer to it. Owing to possible interference from some of the regulatory elements in the topoIIα promoter, however, addition of a strong enhancer might not achieve a high level of basal activity, and could even negate breast cancer specificity (see Figure 4). To obtain a promoter possessing both specificity and a high level of basal activity in breast cancer cells, we reasoned that a minimal breast cancer-specific element linking to an enhancer would have a better chance of success. Thus, we set out to identify a breast cancer-specific element in the topoIIα promoter. To this end, we first explored the possible breast cancer-specific signal to activate topoIIα promoter, then looked for the cis-element that mediated this signaling. TopoIIα expression is upregulated during S phase and peaks during G2/M.37 As the essential cell-cycle regulator Cdk2 is activated during S and G2 phases,38 we asked whether topoIIα promoter activity might be upregulated by Cdk2. We examined the effect of Cdk2 on topoIIα promoter activity by co-transfecting a dominant-negative cdk2 mutant (cdk2-dn)39 and topoIIα promoter reporter construct in a reporter assay. Cdk2-dn suppressed topoIIα promoter activity in three different breast cancer cells but not in normal breast epithelial cells (184A1), lung cancer cells (A549) or nonmalignant hepatocytes (Chang liver) (Figure 2a). As a control, co-transfection of dominant-negative E2F1 mutant had no obvious effect on the topoIIα promoter activity (data not shown). As Cdk2 associates with cyclin E and cyclin A in late G1 and S phases, respectively, we investigated which of these cyclins could activate topoIIα promoter. Expression of cyclin A significantly induced activation of the topoIIα promoter in breast cancer cells, but not in other cell types (Figure 2b). On the other hand, cyclin E had minimal effect on the topoIIα promoter in breast cancer cells (data not shown). These data suggest that cyclin A/Cdk2 activates the topoIIα promoter in a breast cancer-specific manner.
We concluded from these results that the cyclin A/Cdk2-responding element in the topoIIα promoter might be the potential breast cancer-specific element. To identify this element, we generated a series of deletion mutants of the topoIIα promoter, including −182, −90 and −60, which contain three, one and no ICB(s), respectively (Figure 2c).23, 24, 25 Their responses to CCNA activation were examined in the breast cancer cell line SK-BR3 (Figure 2d, left panel). Cyclin A activated the full-length promoter as well as its −182 and −90 deletion mutants, which harbor at least one ICB (−572, −182 and −90 in the left panel of Figure 2d, P<0.05), but not the −60 deletion mutant, which lacks any ICB (−60 in the left panel of Figure 2d). Similar results were obtained when these experiments were performed in two other breast cancer cell lines MDA-MB-468 and MDA-MB-231 (data not shown), indicating that ICBs could be important for mediating the cyclin A signal. This point was further supported by the fact that the mutation of ICB1 in full-length topoIIα promoter resulted in the greater reduction in the response to cyclin A (Figure 2d, right panel), suggesting that ICB1 in the topoIIα promoter is required for full response to cyclin A/Cdk2 activation, and may represent a minimal breast cancer-specific element.
To validate the role of ICB1 in mediating breast cancer specificity of topoIIα promoter, the activity of the −90 deletion mutant, which contained only one ICB site, was examined in a panel of cell lines. Like the full-length promoter (Figure 1a), the −90 promoter had a higher level of activity in most breast cancer cell lines than in other types of cells (Figure 3a), indicating that it retains breast cancer specificity. The ratio of −90 promoter activity to CMV promoter activity (−90/CMV) also had low values in breast cancer cells (<0.5% in Figure 3b). As most other regulatory elements had been deleted in the −90 promoter, we concluded that ICB1 is critical for the breast cancer-specific activity of topoIIα promoter.
Enhancer sequence of the CMV promoter potentiated breast cancer-specific activity of −90 topoIIα promoter in vitro and in vivo
To enhance the basal level of activity of the topoIIα promoter in breast cancer cells, we connected the enhancer sequence of the CMV promoter (CMV enhancer)40 to three ICB-harboring promoters: full-length topoIIα promoter and its −182 and −90 deletion mutants. The composite promoters were designated CT572, CT182 and CT90, respectively (Figure 4a). Interestingly, among these composite promoters, CT90 had the highest level of activity in breast cancer cells relative to that in normal 184A1 cells (Figure 4b, reporter activity shown by log scale). Moreover, the activity of CT90 was comparable to the CMV promoter in breast cancer cells (Figure 4c). As the Y axis represents log scale, Figure 4b showed that the promoter activity of CT90 is about 5- to 10-fold higher than that of CT572 and CT182 in breast cancer cell lines are. CT572 and CT182 had lower specificity for breast cancer cells than CT90 (Figure 4b), and their activities were enhanced to a very limited extent (Figure 4c), suggesting that some elements other than ICB1 in the topoIIα promoter may interfere with CMV enhancer activity.
To test whether CT90 possesses sufficient specificity and activity for systemic liposomal gene therapy, we further examined its activity in vivo. The luciferase constructs driven by the CT90 or CMV promoter (CT90-luc and CMV-luc, respectively) were complexed with DOTAP: Chol liposome, then intravenously injected into mice carrying either MDA-MB-231 breast cancer xenograft. Reporter activity in the normal and tumor tissues was determined by luciferase assay and was shown by the reading per microgram of proteins (Y axis is of log scale, left panels of Figure 4d). The biodistribution of reporter activity from CMV-luc was very similar to that in reported studies: highest in lung, second highest in heart and third highest in liver; it was lower in other tissues. CT90-luc, however, induced higher reporter activity in the tumors and lower activity in normal organs such as heart, lung and liver than CMV-luc did (Figure 4d, left panels), suggesting that CT90 induced less undesired gene expression in normal tissues. It should be mentioned that expression levels in spleen, kidney and muscle are so low and likely negligible. To further evaluate the cancer specificity of the promoters, we calculated the ratio of reporter activity in tumor to that in normal tissue (tumor:normal, right panel, Figure 4d).18 The tumor:normal ratio of the CT90-luc is 49-, 69-, 35-, 4.7-, 4.8- and 9.8-fold higher than that of the CMV-luc in the heart, lung, liver, spleen, kidney and muscle, respectively. These results indicated that the CT90 promoter is more tumor-specific than the CMV promoter. A similar tumor:normal profile of CMV-luc and CT90-luc activities was observed in mice bearing MDA-MB-468 breast cancer xenograft (data not shown). When the liposome–DNA complexes were intratumorally injected, the activity of CT90 promoter was still higher in tumor and lower in normal tissues than that of the CMV promoter (Figure 4e, left panel). In addition, in all the tested tissues, the tumor:normal ratio of CT90 is higher than that of the CMV promoter (Figure 4e, right panel), confirming the higher tumor specificity of CT90. Taken together, these results demonstrate that CT90 possesses higher activity and cancer specificity than the CMV promoter, and may be a better promoter for breast cancer-targeting gene therapy.
CT90-BikDD selectively kills tumor cells in vitro
As CT90 possesses strong breast cancer-specific activity in vitro and in vivo, we compared the therapeutic efficacy of CT90-BikDD and CMV-BikDD constructs, in which the BikDD gene was driven by CT90 and CMV promoter, respectively (Figure 5a). The in vitro luciferase viability assay (see Materials and methods) was performed to examine the cell-killing effects of these constructs in a normal breast epithelial cell line and two breast cancer cell lines (Figure 5b). CT90-BikDD had a more prominent cell-killing effect in breast cancer cells than in normal breast cells, indicating its selectivity for breast cancer cells. On the other hand, the CMV-BikDD treatment killed all cell lines without selectivity.
Liposome-delivered CT90-BikDD selectively suppresses breast tumor growth and prolongs survival in an orthotopic mouse model
To test whether a systemically administered liposome–CT90-BikDD complex could direct selective BikDD expression in vivo, MDA-MB-231 cells were inoculated into the mammary fat pad of female nude mice to form breast tumors. Groups of eight tumor-bearing mice then received weekly intravenous injection of liposomal CT90-BikDD, CMV-BikDD, pGL3 vector (mock treatment) or no treatment (D5W). Tumor growth was suppressed significantly in groups that received CT90-BikDD or CMV-BikDD (P<0.05), compared to that in both mock treatment (pGL3) and no treatment (D5W) groups (Figure 6a). As shown in Figure 6b, the mean survival times of mice in CT90-BikDD, CMV-BikDD, pGL3 and D5W groups were 25.25±1.83, 23.25±1.74, 17.5±1.3 and 16.25±0.98 weeks, respectively. Both CT90-BikDD and CMV-BikDD treatments yielded significant survival benefit for MDA-MB-231 xenograft-bearing mice compared to the no treatment (D5W) or mock treatment (pGL3), indicating that CT90-BikDD provided therapeutic effect comparable to that of CMV-BikDD (Figure 6b, lower panel). The tumor-suppressing effects of liposome–CT90-BikDD complex on other human breast cancer cells were also examined using mice bearing MDA-MB-468 or MDA-MB-435 xenografts, and similar results were observed (data not shown).
We then asked whether the in vivo differential expression profiles of CT90-luc and CMV-luc (Figure 4d) were mirrored by CT90-BikDD and CMV-BikDD. To address this issue, expression of BikDD mRNA in tumor and heart tissues was examined by in situ hybridization (Figure 7). The deep brown staining all over heart tissue from a CMV-BikDD-treated mouse (Figure 7a, upper left panel) indicated a very high expression level of BikDD induced by the CMV promoter, whereas CT90-BikDD treatment induced relatively weak BikDD expression in the heart tissues (Figure 7a, upper right panel). No significant difference in density or level of BikDD expression could be detected between the tumor specimens from mice treated with CT90-BikDD and those treated with CMV-BikDD group (Figure 7b). The negative control using sense BikDD showed no brown staining (Figure 7a and b, lower panels), indicating that the positive signals in the antisense groups came from BikDD expression. These data demonstrate that systemically administered liposome–CT90-BikDD complex can direct selective BikDD expression in breast tumors. Importantly, expression of BikDD in normal organs such as heart was much lower in the CT90-BikDD-treated mice than in the CMV-BikDD-treated mice. Thus, CT90-BikDD possesses antitumor activity comparable to that of CMV-BikDD but may produce fewer side effects because of its lower level of expression in normal tissues.
In gene therapy, the cytotoxicity and immunoginecity of the gene delivery vector, as well as its potential to induce host gene mutations, are critical safety factors. Most of the viral vectors induce a systemic immune response, limiting their use in repeated injections. The presence of viral proteins in normal cells increases the risk of cytotoxicity. Some viral vectors have been reported to induce malignant diseases in clinical trials and preclinical studies41, 42 Compared to viral vectors, liposomes have lower toxicity, lower immunogenicity and almost no risk of inducing host gene mutations. As the goal of this study is to develop a breast cancer-targeting gene therapy regimen for practical application, we adopted a liposome that has been tested in clinical trials, the DOTAP:Chol cationic liposome, as our gene delivery vehicle.
In this study, we identified the breast cancer-specific activity of the topoIIα promoter. We also found that ICBs of the topoIIα promoter respond to cyclin A/Cdk2 in a breast cancer-specific manner. Cyclin A/Cdk2 complex is required during S phase,38 which coincides with increasing topoIIα expression. Cyclin A has been shown to phosphorylate and activate NF-YA,43 the transcription factor that binds to ICBs and activates topoIIα promoter.44 Furthermore, interrupting the association between cyclin A and cdk2 can specifically suppress breast cancer cell growth.45, 46 Together with our results, these data indicate that cyclin A/Cdk2 is a specific upstream activator of topoIIα promoter in breast cancer cells. Interestingly, among the deletion mutants of topoIIα promoter, only the minimal −90 promoter was potentiated successfully by the CMV enhancer, resulting in activity comparable to that of the CMV promoter while retaining specificity for breast cancer. This demonstrated that other regulatory sites in the upstream region of the topoIIα promoters may have adverse interaction with the enhancer sequence.
Tumor vasculature is usually irregular and has dilated lumens and fragile basement membranes, forming a leaky structure.47, 48, 49 Increased permeability would facilitate movement of liposome complex from vessels into tumor interstitium. Actually, it has been reported that cationic liposome complex is preferentially accumulated in tissues with high density of capillaries (e.g. thyroid or ovary), chronic inflammation and tumors.50 When the luciferase gene is delivered by liposome through tail vein injection in a mouse model, however, the principal sites of luciferase activity are the lung, heart and liver, and the luciferase activity in lung per microgram of tissue protein is one to two order higher than that in any other organs.14, 15, 16, 17 This fact could reflect that gene delivery efficiency depends not only on the interstitial retention, but also on the uptake of tissues, nuclear transport of DNA and tissue-specific promoter activity.51 In this study, a similar biodistribution pattern was observed when the DOTAP: Chol–CMV-luc complex was injected through the tail vein of tumor-bearing mice (Figure 4d, upper panels). However, CT90-luc induced a much lower luciferase activity in normal organs, especially lung and heart, and higher levels of activity in tumors than CMV-luc. Therefore, CT90 fits our two criteria for a promoter in systemic cancer gene therapy described in Introduction.
Proapoptotic genes can kill cells efficiently regardless of their proliferation rate; thus, they are good candidate therapeutic genes for cancer gene therapy. bik is a BH3-only proapoptotic gene that has been recognized as an essential initiator of apoptosis.52, 53 Recently, we found that mutation of threonine 33 and serine 35 of Bik to aspartic acid increases its apoptotic activity in vitro and its antitumor activity in animal models. The Bik mutant BikDD is more potent in inducing apoptosis than wild-type Bik.26 In our mouse models of orthotopic breast tumor xenografts, treatment with CT90-BikDD or CMV-BikDD yielded comparable efficacies in tumor growth suppression (Figure 6a) and survival benefits (Figure 6b). More importantly, the CT90 promoter drives much lower levels of gene expression than the CMV promoter in normal organs such as heart (Figure 7a), indicating that it may reduce the risk of side effects of therapy. As CMV promoter-driven constructs have been used widely in clinical trials of cancer gene therapy (according to The NIH/FDA Genetic Modification Clinical Research Information System (GeMCRIS; www4.od.nih.gov/oba/rac/GeMCRIS/GeMCRIS.htm) and The Journal of Gene Medicine Clinical Trial site (www.wiley.co.uk/genetherapy/clinical/)), the potential of the CT90 promoter to reduce side effects of cancer gene therapy has very important clinical implications. In gene therapy protocols using a potent proapoptotic gene, CT90 may be a better choice of promoter than CMV for both therapeutic efficacy and safety.
In summary, we identified the breast cancer-specific activity of the topoIIα promoter and its activating signal. On the basis of this finding, we have generated a CT90 promoter by linking topoIIα with an enhancer element of CMV promoter, and used it to develop a systemic breast cancer gene therapy delivered by liposome–DOTAP: Chol–CT90-BikDD. Systemically administered liposome–CT90-BikDD complex can direct BikDD expression in breast cancer cells in vivo, effectively suppress tumor growth and prolong the survival of tumor-bearing mice. This preclinical study shows that liposome–CT90-BikDD complex has a potential to be developed further for breast cancer treatment.
Weir HK, Thun MJ, Hankey BF, Ries LA, Howe HL, Winngo PA et al. Annual report to the nation on the status of cancer, 1975–2000, featuring the uses of surveillance data for cancer prevention and control. J Natl Cancer Inst 2003; 95: 1276–1299.
Rouse J, Jackson SP . Interfaces between the detection, signaling, and repair of DNA damage. Science 2002; 297: 547–551.
Kolodner RD, Putnam CD, Myung K . Maintenance of genome stability in Saccharomyces cerevisiae. Science 2002; 297: 552–557.
Maser RS, DePinho RA . Connecting chromosomes, crisis, and cancer. Science 2002; 297: 565–569.
Hortobagyi GN, Ueno NT, Xia W, Zhang S, Wolf JK, Putnam JB et al. Cationic liposome-mediated E1A gene transfer to human breast and ovarian cancer cells and its biologic effects: a phase I clinical trial. J Clin Oncol 2001; 19: 3422–3433.
Ueno NT, Bartholomeusz C, Xia W, Anklesaria P, Bruckheimer EM, Mebel E et al. Systemic gene therapy in human xenograft tumor models by liposomal delivery of the E1A gene. Cancer Res 2002; 62: 6712–6716.
Yoo GH, Hung MC, Lopez-Berestein G, LaFollete S, Ensley JF, Carey M et al. Phase I trial of intratumoral liposome E1A gene therapy in patients with recurrent breast and head and neck cancer. Clin Cancer Res 2001; 7: 1237–1245.
Dummer R, Bergh J, Karlsson Y, Horowitz JA, Mulder NH, Huinink DTB et al. Biological activity and safety of adenoviral vector-expressed wild-type p53 after intratumoral injection in melanoma and breast cancer patients with p53-overexpressing tumors. Cancer Gene Ther 2000; 7: 1069–1076.
Pandha HS, Martin LA, Rigg A, Hurst HC, Stamp GW, Sikora K et al. Genetic prodrug activation therapy for breast cancer: a phase I clinical trial of erbB-2-directed suicide gene expression. J Clin Oncol 1999; 17: 2180–2189.
Saukkonen K, Hemminki A . Tissue-specific promoters for cancer gene therapy. Expert Opin Biol Ther 2004; 4: 683–696.
Brand K, Loser P, Arnold W, Bartels T, Strauss M . Tumor cell-specific transgene expression prevents liver toxicity of the adeno-HSVtk/GCV approach. Gene Therapy 1998; 5: 1363–1371.
Ikegami S, Tadakuma T, Ono T, Suzuki S, Yoshimura I, Asano T et al. Treatment efficiency of a suicide gene therapy using prostate-specific membrane antigen promoter/enhancer in a castrated mouse model of prostate cancer. Cancer Sci 2004; 95: 367–370.
Haviv YS, Curiel DT . Conditional gene targeting for cancer gene therapy. Adv Drug Deliv Rev 2001; 53: 135–154.
Lu H, Zhang Y, Roberts DD, Osborne CK, Templeton NS . Enhanced gene expression in breast cancer cells in vitro and tumors in vivo. Mol Ther 2002; 6: 783–792.
Templeton NS, Lasic DD, Frederik PM, Strey HH, Roberts DD, Pavlakis GN . Improved DNA: liposome complexes for increased systemic delivery and gene expression. Nat Biotechnol 1997; 15: 647–652.
Barron LG, Szoka Jr FC . The perplexing delivery mechanism of lipoplexes. In: Huang L, Hung MC, Wagner E (eds). Non-viral Vectors for Gene Therapy. Academic Press: San Diego, CA, 1999, pp 229–266.
Li S, Rizzo MA, Bhattacharya S, Huang L . Characterization of cationic lipid–protamine–DNA (LPD) complexes for intravenous gene delivery. Gene Therapy 1998; 5: 930–937.
Zou Y, Peng H, Zhou B, Wen Y, Wang SC, Tsai EM et al. Systemic tumor suppression by the proapoptotic gene bik. Cancer Res 2002; 62: 8–12.
Anderson LM, Krotz S, Weitzman SA, Thimmapaya B . Breast cancer-specific expression of the Candida albicans cytosine deaminase gene using a transcriptional targeting approach. Cancer Gene Ther 2000; 7: 845–852.
Katabi MM, Chan HL, Karp SE, Batist G . Hexokinase type II: a novel tumor-specific promoter for gene-targeted therapy differentially expressed and regulated in human cancer cells. Hum Gene Ther 1999; 10: 155–164.
Lin T, Huang X, Gu J, Zhang L, Roth JA, Xiong M et al. Long-term tumor-free survival from treatment with the GFP-TRAIL fusion gene expressed from the hTERT promoter in breast cancer cells. Oncogene 2002; 21: 8020–8028.
Hung MC, Hortobagyi GN, Ueno NT . Development of clinical trial of E1A gene therapy targeting HER-2/neu-overexpressing breast and ovarian cancer. Adv Exp Med Biol 2000; 465: 171–180.
Adachi N, Nomoto M, Kohno K, Koyama H . Cell-cycle regulation of the DNA topoisomerase IIalpha promoter is mediated by proximal CCAAT boxes: possible involvement of acetylation. Gene 2000; 245: 49–57.
Falck J, Jensen PB, Sehested M . Evidence for repressional role of an inverted CCAAT box in cell cycle-dependent transcription of the human DNA topoisomerase IIalpha gene. J Biol Chem 1999; 274: 18753–18758.
Hochhauser D, Stanway CA, Harris AL, Hickson ID . Cloning and characterization of the 5′-flanking region of the human topoisomerase II alpha gene. J Biol Chem 1992; 267: 18961–18965.
Li YM, Wen Y, Zhou BP, Kuo HP, Ding Q, Hung MC . Enhancement of Bik antitumor effect by Bik mutants. Cancer Res 2003; 63: 7630–7633.
Kurachi S, Hitomi Y, Furukawa M, Kurachi K . Role of intron I in expression of the human factor IX gene. J Biol Chem 1995; 270: 5276–5281.
Xu X, Lotan R . Nonisotopic in situ hybridization for the detection of nuclear retinoid receptor transcripts in tissue sections. Methods Mol Biol 1998; 89: 233–246.
Sitzmann J, Noben-Trauth K, Kamano H, Klempnauer KH . Expression of B-Myb during mouse embryogenesis. Oncogene 1996; 12: 1889–1894.
Sotiriou C, Neo SY, McShane LM, Korn EL, Long PM, Jazaeri A et al. Breast cancer classification and prognosis based on gene expression profiles from a population-based study. Proc Natl Acad Sci USA 2003; 100: 10393–10398.
van ’t Veer LJ, Dai H, van de Vijver MJ, He JD, Hart AA, Bernards R et al. Gene expression profiling predicts clinical outcome of breast cancer. Nature 2002; 415: 530–536.
Sorlie T, Perou CM, Tibshirani R, Aas T, Geisler S, Johnsen H et al. Gene expression patterns of breast carcinomas distinguish tumor subclasses with clinical implications. Proc Natl Acad Sci USA 2001; 98: 10869–10874.
van de Vijver MJ, He YD, van't Veer LJ, Dai H, Hart AA, Voskuil DW et al. A gene-expression signature as a predictor of survival in breast cancer. N Engl J Med 2002; 347: 1999–2009.
Nacht M, Ferguson AT, Zhang W, Petroziello JM, Cook BP, Gao YH et al. Combining serial analysis of gene expression and array technologies to identify genes differentially expressed in breast cancer. Cancer Res 1999; 59: 5464–5470.
Leerkes MR, Caballero OL, Mackay A, Torloni H, O'Hare MJ, Simpson AJ et al. In silico comparison of the transcriptome derived from purified normal breast cells and breast tumor cell lines reveals candidate upregulated genes in breast tumor cells. Genomics 2002; 79: 257–265.
Castagnetta LA, Agostara B, Montalto G, Polito L, Campisi I, Saetta A et al. Local estrogen formation by nontumoral, cirrhotic, and malignant human liver tissues and cells. Cancer Res 2003; 63: 5041–5045.
Woessner RD, Mattern MR, Mirabelli CK, Johnson RK, Drake FH . Proliferation- and cell cycle-dependent differences in expression of the 170 kilodalton and 180 kilodalton forms of topoisomerase II in NIH-3T3 cells. Cell Growth Differ 1991; 2: 209–214.
Vermeulen K, Van Bockstaele DR, Berneman ZN . The cell cycle: a review of regulation, deregulation and therapeutic targets in cancer. Cell Prolif 2003; 36: 131–149.
van den Heuvel S, Harlow E . Distinct roles for cyclin-dependent kinases in cell cycle control. Science 1993; 262: 2050–2054.
Xu L, Daly T, Gao C, Flotte TR, Song S, Byrne BJ et al. CMV-beta-actin promoter directs higher expression from an adeno-associated viral vector in the liver than the cytomegalovirus or elongation factor 1 alpha promoter and results in therapeutic levels of human factor X in mice. Hum Gene Ther 2001; 12: 563–573.
Yi Y, Hahm SH, Lee KH . Retroviral gene therapy: safety issues and possible solutions. Curr Gene Ther 2005; 5: 25–35.
Marshall E . Gene therapy. Viral vectors still pack surprises. Science 2001; 294: 1640.
Yun J, Chae HD, Choi TS, Kim EH, Bang YJ, Chung J et al. Cdk2-dependent phosphorylation of the NF-Y transcription factor and its involvement in the p53–p21 signaling pathway. J Biol Chem 2003; 278: 36966–36972.
Joshi AA, Wu Z, Reed RF, Suttle DP . Nuclear factor-Y binding to the topoisomerase IIalpha promoter is inhibited by both the p53 tumor suppressor and anticancer drugs. Mol Pharmacol 2003; 63: 359–367.
Mendoza N, Fong S, Marsters J, Koeppen H, Schwall R, Wickramasinghe D . Selective cyclin-dependent kinase 2/cyclin A antagonists that differ from ATP site inhibitors block tumor growth. Cancer Res 2003; 63: 1020–1024.
Chen YN, Sharma SK, Ramsey TM, Jiang L, Martin MS, Baker K et al. Selective killing of transformed cells by cyclin/cyclin-dependent kinase 2 antagonists. Proc Natl Acad Sci USA 1999; 96: 4325–4329.
Baillie CT, Winslet MC, Bradley NJ . Tumour vasculature – a potential therapeutic target. Br J Cancer 1995; 72: 257–267.
Kan Z, Ivancev K, Lunderquist A, McCuskey PA, McCuskey RS, Wallace S . In vivo microscopy of hepatic tumors in animal models: a dynamic investigation of blood supply to hepatic metastases. Radiology 1993; 187: 621–626.
Liotta LA, Saidel MG, Kleinerman J . The significance of hematogenous tumor cell clumps in the metastatic process. Cancer Res 1976; 36: 889–894.
Thurston G, McLean JW, Rizen M, Baluk P, Haskell A, Murphy TJ et al. Cationic liposomes target angiogenic endothelial cells in tumors and chronic inflammation in mice. J Clin Invest 1998; 101: 1401–1413.
McLean JW, Thurston G, McDonald DM . Sites of uptake and expression of cationic liposome/DNA complexes injected intravenously. In: Huang L, Hung M-C, Wagner E (eds). Nonviral Vectors for Gene Therapy. Academic Press: San Diego, CA, 1999, pp 120–134.
Boyd JM, Gallo GJ, Elangovan B, Houghton AB, Malstrom S, Avery BJ et al. Bik, a novel death-inducing protein shares a distinct sequence motif with Bcl-2 family proteins and interacts with viral and cellular survival-promoting proteins. Oncogene 1995; 11: 1921–1928.
Han J, Sabbatini P, White E . Induction of apoptosis by human Nbk/Bik, a BH3-containing protein that interacts with E1B 19K. Mol Cell Biol 1996; 16: 5857–5864.
We thank Ms Chu-Li Weng for the technical assistance. C-PD is an awardee of Department of Defense Breast Cancer Research Project Training Grant. K-M Rau is a visiting scientist from Chang Gung Memorial Hospital, Kaohsiung Hsien, Taiwan. This work was supported by Susan G Komen Breast Cancer Foundation Research Grant, US Army Breast Cancer Research Program Center of Excellence, Breast Cancer Research Foundation and MDACC Cancer Center Supporting Grant (CA16672).
About this article
Cite this article
Day, C., Rau, K., Qiu, L. et al. Mutant Bik expression mediated by the enhanced minimal topoisomerase IIα promoter selectively suppressed breast tumors in an animal model. Cancer Gene Ther 13, 706–719 (2006). https://doi.org/10.1038/sj.cgt.7700945
- breast cancer
- topoisomerase IIα promoter