DNA methyltransferase inhibitor, zebularine, delays tumor growth and induces apoptosis in a genetically engineered mouse model of breast cancer

Abstract

Zebularine is a novel potent inhibitor of both cytidine deaminase and DNA methylation. We examined the effect of zebularine on mammary tumor growth in genetically engineered MMTV-PyMT transgenic mice that develop mammary tumors at 60 days of age with 100% penetrance. The MMTV-PyMT transgenic mice were randomized at 46 days of age into control (n = 25) and zebularine (n = 25) treatment groups and monitored for parameters of tumor growth. Zebularine was administered at 5 mg/mL in drinking water. We observed a significant delay in the growth of mammary tumors in zebularine-treated mice with a statistically significant reduction (P = 0.0135) in total tumor burden at 94 days of age when the mice were sacrificed. After 48 days of zebularine treatment, the tumors were predominantly necrotic compared with untreated animals. In addition, a high apoptotic index by terminal deoxynucleotidyl transferase-mediated dUTP nick end labeling assay was observed as early as 13 days following treatment. Immunoblot analysis showed depletion of DNMT1 and partial depletion of DNMT3b after zebularine treatment. Microarray analyses of global gene expression identified upregulation of twelve methylation-regulated genes as well as a set of candidate cancer genes that participate in cell growth and apoptosis. In summary, zebularine inhibits the growth of spontaneous mammary tumors and causes early onset of tumor cell necrosis and apoptosis in a genetically engineered mouse model of breast cancer. Defining the parameters of zebularine-mediated tumor inhibition may advance the future development of DNA methyltransferase inhibitors as an effective cancer treatment.

Figure 1.

Effect of zebularine on tumor growth in MMTV-PyMT transgenic mice. A, chemical structure of zebularine. B, schematic representation of zebularine treatment. Cohorts of female MMTV-PyMT mice (n = 50) were monitored for onset of palpable mammary tumors and size of the tumor. At 46 days of age, the mice were randomized into control and zebularine-treated groups (n = 25 each). For zebularine treatment, mice were treated via solubilizing 5 mg/mL zebularine in sucrose-supplemented water. Control mice received sucrose-supplemented drinking water. C and D, delay of tumor growth in zebularine-treated mice. C, the average tumor volume for both axillary and the inguinal tumors were calculated for zebularine-treated (n = 10) or control (n = 10) mice. The data for the first axillary and first inguinal tumors is shown. D, the average tumor volume for axillary tumors was calculated for zebularine-treated (n = 10) or control (n = 10) mice. Only the data for the first axillary tumor detected in each animal are shown. E, the average tumor volume for inguinal tumors were calculated for zebularine-treated (n = 10) or control (n = 10) mice. Only the data for the first inguinal tumor detected are shown. Triangles and circles represent mean tumor volume for control and zebularine-treated mice, respectively. Bars represent SD. F, ten mice in each group were treated to 94 days of age at which time they were sacrificed, and the total tumor burden was excised and weighed and presented as mean ± SD. The total tumor burden for the control and treated animals was compared by Student t test. P values are shown in C to F.

Figure 2.

Effect of zebularine treatment on tumor necrosis and apoptosis in MMTV-PyMT transgenic mice. A, extensive necrosis characterized by acellular pink area was detected in 50% of section after 48 days of zebularine treatment (a and c). Negligible necrosis in control animals (b and d). Magnification, ×200. B, TUNEL assay revealed extensive apoptotic cells in mammary tumors after 13 and 48 days of zebularine treatment (b and d). Negligible level of apoptotic cells detected in untreated tumors (a and c). C, quantitative analysis of apoptotic cells with Automated Cellular Imaging System shows a higher apoptotic index after 13 (n = 4 mice) and 48 days (n = 3 mice) of zebularine treatment compared with the untreated tumor. P value is indicated. Zeb, zebularine.

Figure 3.

Depletion of DNMT1 and DNMT3b in zebularine-treated transgenic mice. A, immunoblot analysis of DNMT1. Tumors from both untreated (n = 3; lanes 1–3) and zebularine-treated mice (n = 3; lanes 4–6) were isolated after 23 days of treatment. B, immunoblot analysis of DNMT3b from untreated tumors (n = 3; lanes 1–3) and zebularine-treated mice (n = 3; lanes 4–6). C, immunoblot analysis of DNMT1 and DNMT3b from lysates isolated from normal mammary gland 4 (lanes 1–3) as compared with additional untreated tumor specimens (lanes 4 and 5). α-Tubulin was used as a protein loading control. ΜG, mammary gland.

Figure 4.

Validation of microarray analysis of zebularine-regulated gene expression by real-time RT-PCR. A, comparison of the relative fold change of zebularine-regulated genes as measured by microarray or quantitative real-time RT-PCR. B, positive correlation between real-time RT-PCR and microarray analysis of zebularine-regulated genes (R² = 0.94).

Figure 5.

The mRNA expression level and methylation status of selected zebularine-activated genes are altered during mammary tumor development and zebularine treatment. A, quantitative RT-PCR measurement of Twist2, Becn1, Sfrp1, Cdkn1a, and H19 expression in normal mammary glands 2 and 3 and 4 measured independently from 70-day-old FVB mice, mammary tumors from 69-day-old and 94-day-old PyMT transgenic mice, and tumor samples from zebularine-treated PyMT mice [69-day for all genes except 94-day for Twist2; 69 and 94 days corresponds to 23 and 48 days after zebularine (Zeb) treatment, respectively]. Each bar represents the mean gene expression level measured in 3 individual mice. B, the architecture of the mouse Twist2 locus. Included is the GC percentage and location of exon 1 within this locus, the transcriptional start site (ATG), location of PCR primers used to amplify a portion of the Twist2 gene, and the gene sequence and individual CpG dinucleotides interrogated by pyrosequencing. C, The percentage of methylated CpG dinucleotides within the Twist2 gene in indicated tissue or tumor samples measured by pyrosequencing. Measurements in normal mammary glands 2 and 3 and 4 from control FVB mice are graphed separately. Each bar indicates mean methylation of the indicated CpG dinucleotide within the Twist2 gene, measured in 3 independent tissue or tumor samples (error bars, ± SE). Also graphed is the average CpG methylation at all 4 CpG sites analyzed. Statistical test used was a 2-tailed t test comparing normal mammary glands or zebularine-treated tumors to untreated tumors (*, P < 0.05).