Triplex-forming oligonucleotides targeting c-MYC potentiate the anti-tumor activity of gemcitabine in a mouse model of human cancer

Abstract

Antimetabolite chemotherapy remains an essential cancer treatment modality, but often produces only marginal benefit due to the lack of tumor specificity, the development of drug resistance, and the refractoriness of slowly proliferating cells in solid tumors. Here, we report a novel strategy to circumvent the proliferation-dependence of traditional antimetabolite-based therapies. Triplex-forming oligonucleotides (TFOs) were used to target site-specific DNA damage to the human c-MYC oncogene, thereby inducing replication-independent, unscheduled DNA repair synthesis (UDS) preferentially in the TFO-targeted region. The TFO-directed UDS facilitated incorporation of the antimetabolite, gemcitabine (GEM), into the damaged oncogene, thereby potentiating the anti-tumor activity of GEM. Mice bearing COLO 320DM human colon cancer xenografts (containing amplified c-MYC) were treated with a TFO targeted to c-MYC in combination with GEM. Tumor growth inhibition produced by the combination was significantly greater than with either TFO or GEM alone. Specific TFO binding to the genomic c-MYC gene was demonstrated, and TFO-induced DNA damage was confirmed by NBS1 accumulation, supporting a mechanism of enhanced efficacy of GEM via TFO-targeted DNA damage-induced UDS. Thus, coupling antimetabolite chemotherapeutics with a strategy that facilitates selective targeting of cells containing amplification of cancer-relevant genes can improve their activity against solid tumors, while possibly minimizing host toxicity.

Keywords: DNA-reactive agents; combination chemotherapy; oncogenes; triplex-forming oligonucleotides; xenograft models.

Figure 1

Specific binding of Myc2T to genomic DNA in vitro and in vivo. (A) Schematic of the Myc2T binding domain. The “X” indicates the psoralen-crosslinking site, and arrows represent PCR primers. (B) Binding assays were performed using radiolabeled TFO Myc2T (lanes 1–7) or control oligonucleotide TFOc (lanes 8–14). (C) pMyc2T (lanes 3 and 4), and control pTFOc (lanes 1 and 2) were incubated with COLO 320DM genomic DNA (gDNA) followed by UVA irradiation (lanes 2 and 4). (D) pTFOc-bio (lanes 1 and 3) and pMyc2T-bio (lanes 2 and 4) were administered passively (lanes 1 and 2) or with a transfection reagent (lanes 3 and 4) into COLO 320DM cells. DNA was analyzed on a native polyacrylamide gel. (E) DNA from Figure 1D was PCR amplified, and the products were visualized on an agarose gel.

Figure 2

TFOs induce DNA damage and GEM incorporation at the TFO target site. (A) Schematic of the TFO-targeted region with two different primer sets, 2 bp and 4 kb from the triplex-forming site. (B) A representative agarose gel of a ChIP assay performed with extracts of TFO-treated COLO 320DM cells using primer binding sites 2 bp (lanes 2–5) and 4 kb (lanes 7–10) from the triplex-forming site. (C) Fold enrichment of NBS1 binding in the 2 bp primer set region adjacent to the Myc2T binding site of Myc2T-treated cells compared to TFOc-treated cells, normalized to that in the 4 kb primer set (*P<0.05 as calculated by one sample Student’s t-test against 1). (D) COLO 320DM cells were incubated with [³H]GEM and either Myc2T or control TFOc. The bar graph illustrates the fold increase of [³H]GEM of the TFO-targeted 200–700 bp fragment compared to a background sample of ~5 kb. (*P<0.05, Student’s t-test, unequal variance, for all samples against GEM + Myc2T, error bars indicate standard deviation).

Figure 3

TFO enhanced antitumor activity of GEM against human colon carcinoma xenografts in mice. (A) Average tumor volume versus days post-surgical implantation. Mice were treated with PBS, Myc2T, TFOc, gemcitabine at 30 mg/kg of body weight (Gem30), at 50 mg/kg (Gem50), or combinations of oligonucleotides and gemcitabine. The mean tumor volume for each treatment group is plotted versus time. Gem50+Myc2T significantly inhibited tumor growth compared to Gem50 (P<0.05) and Gem50+TFOc (P<0.07). Each group consisted of at least 3 mice. (B) The average time for tumors to reach a volume of 0.85 cm³ is shown. Error bars indicate standard deviations from 3 independent experiments. The amount of time required for Gem50+Myc2T treated tumors to reach a volume of 0.85 cm³ was significantly longer than that of all other groups when analyzed by using the Student’s t-test, unequal variance, 2-sided (P<0.05).

Figure 4

Proposed mechanism for TFO-induced potentiation of GEM antitumor efficacy. The TFO binds its target sequence resulting in DNA strand breaks that stimulate repair synthesis, thereby facilitating incorporation of GEM into the TFO-targeted site, leading to enhanced GEM incorporation at the targeted oncogene and increased cell death.