Low doses of selenium specifically stimulate the repair of oxidative DNA damage in LNCaP prostate cancer cells

Abstract

Epidemiological studies have demonstrated an inverse relationship between selenium (Se) intake and cancer incidence and/or mortality. However, the molecular mechanisms underlying the cancer chemopreventive activity of Se compounds remain largely unknown. The objective of this study was to investigate the effect of low doses of Se on the stimulation of DNA repair systems in response to four different qualities of DNA damage. P53-proficient LNCaP human prostate adenocarcinoma cells were grown either untreated or in the presence of low concentrations of two Se compounds (30° nM sodium selenite, or 10 μM selenomethionine) and exposed to UVA, H2O2, methylmethane sulfonate (MMS) or UVC. Cell viability as well as DNA damage induction and repair were evaluated by the alkaline Comet assay. Overall, Se was shown to be a very potent protector against cell toxicity and genotoxicity induced by oxidative stress (UVA or H2O2) but not from the agents that induce other types of deleterious lesions (MMS or UVC). Furthermore, Se-treated cells exhibited increased oxidative DNA repair activity, indicating a novel mechanism of Se action. Therefore, the benefits of Se could be explained by a combination of antioxidant activity, the reduction in DNA damage and the enhancement of oxidative DNA repair capacity.

Figure 1

Se supplementation increases resistance to oxidative stress. Cell viability was determined by the MTT assay 24 h after exposure and data were presented relative to the control cells. LNCaP cells were pre-treated or post-treated with 30 nM SS or 10 μM SM for 72 h or 24 h. The doses of Se used for media supplementation was chosen in order to optimize GPx1 activity in LNCaP cells. (A) Cells (30 min on ice). (C) Cells were irradiated with increasing doses of UVA. (B) Cells were incubated with several concentrations of H₂O₂ were incubated with several concentrations of MMS (15 min at 37°C). (D) Cells were irradiated with increasing doses of UVC. Values are given as mean ± SD of n = 3 independent experiments and triplicate measurements, and we used for statistical significance using Student’s t-test. *p < 0.05 and **p < 0.01 vs. NT.

Figure 2

Se supplementation increases long-term resistance to oxidative stress. Cell viability was assessed by the clonogenic assay (or colony formation assay) which is an in vitro cell survival assay based on the ability of a single cell to grow into a colony after two weeks. LNCaP cells were plated in 35 mm petri dishes and incubated with SS or SM for 72 h at 37°C and then irradiated with growing dose of (A) UVA or (B) incubated with several concentrations of H₂O₂ (30 min on ice) and immediately re-plated in 60 mm petri dishes at a density of 50 cells/mL for 12 days. Cells were then fixed with ethanol, stained with crystal violet in order to count the number of colonies. Results are expressed as the mean percentage ± SD of colonies of 3 independent experiments and triplicate measurements, and we used for statistical significance using Student’s t-test. *p < 0.05 and **p < 0.01 vs. NT.

Figure 3

Se protects against oxidative DNA damage, assessed by alkaline comet assay after UVA, H₂O₂, MMS and UVC-treated LNCaP cells. To evaluate the extent of DNA damage by the Comet assay, LNCaP cells (50,000 cells/Petri dish 60 mm) were pre-treated with 30 nM SS or 10 μM SM for 72 h at 37°C and exposed to either 50 J/cm² UVA, 200 μM H₂O₂ 10 J/m² UVC, or 500 μM MMS. Control conditions were NT, SS and SM cells sham-treated with PBS. Immediately after treatment, cells were collected and lysed in order to evaluate DNA damage following Comet assay protocol. (A) Typical images of comets in UVA-treated LNCaP cells and effects of Se supplementation are presented. LNCaP cell line was cultured with or without 30 nm SS or 10 μM SM for 72 h; each group was treated (200 μM for 30 min on ice), (C) MMS (500 μM for 15 min at 37°C) and (D) UVC with a specific dose of (B) UVA (50 J/cm²), H₂O₂ (10 J/m²). Immediately after stress, cells were collected and comet assay was performed. Values are given as mean ± SD of n = 3 independent experiments and triplicate measurements. The mean of the tail intensity (n = 3) was calculated and analyzed for statistical significance using Student’s t-test. (*p < 0.05 vs NT UVA and NT H₂O₂; Δp < 0.05 vs NT UVA Fpg and NT H₂O₂ Fpg).

Figure 4

Dose-response curve for the formation of 8-oxoGua after UVA irradiation detected by HPLC-MS/MS analysis. HPLC coupled with tandem mass spectrometry was used to quantify 8-oxoGua lesion amount in LNCaP cells after UVA exposure. Cells were pre-treated with SS or SM for 72 h and then irradiated with growing dose of UVA (0, 50, 100 and 200 J/cm ²). Immediately after irradiation cells were collected and DNA samples were extracted and hydrolyzed. We observed that cells pre-treated with SM significantly reduced 8-oxoGua formation compared to other conditions (NT and SS). Values are given as mean of 8-oxoGua per 10⁶ normal bases ± SD of n = 3 independent experiments and triplicate measurements and analyzed for statistical significance using Student’s t-test. *p < 0.05 vs. NT.

Figure 5

Se supplementation increases 8oxoGua excision capacity of cell extracts. Comet assay-based approach was used to evaluate the activity of several glycosylases involved in the recognition and excision of the lesions induced on our genomic substrates. Total protein cell extracts (NT, SS, and SM) were prepared and their excision capacities were tested into different DNA damaged substrate during 30 min at 37°C, using a modified version of comet assay. Three different DNA damaged substrate were used (A) 1 μM Riboflavin + 10 J/cm² UVA, (B) 500 μM MMS, (C) 10 J/m ² UVC. For each substrate we used an undamaged substrate as a control. Incubation with several repair enzymes was used as positive control (Fpg, Endo III, and T4 Endo V). The SS and SM protein extracts are more efficient for the excision of the lesion generated on substrate (A) compared to (B) and (C). This oxidative substrate is enriched with 8-oxoGua. Three biological replicates were tested in triplicate. The mean of % of excision of each cell extract (n = 3) was calculated and statistical significance was assessed using Student’s t-test. (p < 0.05 vs. buffer; *p < 0.05 vs. NT cell extract, ^# p < 0.05 cell extract).