Radiation-induced salivary gland dysfunction results from p53-dependent apoptosis

Abstract

Purpose: Radiotherapy for head-and-neck cancer causes adverse secondary side effects in the salivary glands and results in diminished quality of life for the patient. A previous in vivo study in parotid salivary glands demonstrated that targeted head-and-neck irradiation resulted in marked increases in phosphorylated p53 (serine(18)) and apoptosis, which was suppressed in transgenic mice expressing a constitutively active mutant of Akt1 (myr-Akt1).

Methods and materials: Transgenic and knockout mouse models were exposed to irradiation, and p53-mediated transcription, apoptosis, and salivary gland dysfunction were analyzed.

Results: The proapoptotic p53 target genes PUMA and Bax were induced in parotid salivary glands of mice at early time points after therapeutic radiation. This dose-dependent induction requires expression of p53 because no radiation-induced expression of PUMA and Bax was observed in p53-/- mice. Radiation also induced apoptosis in the parotid gland in a dose-dependent manner, which was p53 dependent. Furthermore, expression of p53 was required for the acute and chronic loss of salivary function after irradiation. In contrast, apoptosis was not induced in p53-/- mice, and their salivary function was preserved after radiation exposure.

Conclusions: Apoptosis in the salivary glands after therapeutic head-and-neck irradiation is mediated by p53 and corresponds to salivary gland dysfunction in vivo.

Figure 1. Induction of p53-dependent apoptotic genes in FVB wildtype and myr-Akt1 transgenic mice following radiation in vivo

The head and neck region of FVB wildtype and myr-Akt1 transgenic mice was exposed to 0, 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 or 8 hours post-irradiation. PUMA or Bax expression was graphed using the 2^−ΔΔ C_t method and normalized to untreated wildtype. Induction of PUMA in FVB wildtype and myr-Akt transgenic mice at 4 (A) or 8 (B) hours. Induction of Bax in FVB wildtype and myr-Akt transgenic mice at 4 (C) or 8 (D) hours. Significant differences (p≤0.05) were determined using an ANOVA followed by a Bonferroni test adjusted for the number of pairwise comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 1. Induction of p53-dependent apoptotic genes in FVB wildtype and myr-Akt1 transgenic mice following radiation in vivo

The head and neck region of FVB wildtype and myr-Akt1 transgenic mice was exposed to 0, 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 or 8 hours post-irradiation. PUMA or Bax expression was graphed using the 2^−ΔΔ C_t method and normalized to untreated wildtype. Induction of PUMA in FVB wildtype and myr-Akt transgenic mice at 4 (A) or 8 (B) hours. Induction of Bax in FVB wildtype and myr-Akt transgenic mice at 4 (C) or 8 (D) hours. Significant differences (p≤0.05) were determined using an ANOVA followed by a Bonferroni test adjusted for the number of pairwise comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 1. Induction of p53-dependent apoptotic genes in FVB wildtype and myr-Akt1 transgenic mice following radiation in vivo

The head and neck region of FVB wildtype and myr-Akt1 transgenic mice was exposed to 0, 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 or 8 hours post-irradiation. PUMA or Bax expression was graphed using the 2^−ΔΔ C_t method and normalized to untreated wildtype. Induction of PUMA in FVB wildtype and myr-Akt transgenic mice at 4 (A) or 8 (B) hours. Induction of Bax in FVB wildtype and myr-Akt transgenic mice at 4 (C) or 8 (D) hours. Significant differences (p≤0.05) were determined using an ANOVA followed by a Bonferroni test adjusted for the number of pairwise comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 1. Induction of p53-dependent apoptotic genes in FVB wildtype and myr-Akt1 transgenic mice following radiation in vivo

The head and neck region of FVB wildtype and myr-Akt1 transgenic mice was exposed to 0, 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 or 8 hours post-irradiation. PUMA or Bax expression was graphed using the 2^−ΔΔ C_t method and normalized to untreated wildtype. Induction of PUMA in FVB wildtype and myr-Akt transgenic mice at 4 (A) or 8 (B) hours. Induction of Bax in FVB wildtype and myr-Akt transgenic mice at 4 (C) or 8 (D) hours. Significant differences (p≤0.05) were determined using an ANOVA followed by a Bonferroni test adjusted for the number of pairwise comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 2. Induction of p53-dependent apoptotic genes in p53 mice following irradiation in vivo

The head and neck region of p53+/+, p53+/−, and p53−/− mice was exposed to 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 hours post-irradiation. PUMA and Bax expression was analyzed as described in Figure 1. Induction of PUMA (A) and Bax (B) in p53 genotype mice. Significant differences (p≤0.05) were determined using an ANOVA followed by a pairwise Bonferroni test adjusting for the number of comparisons. (*) indicates significant difference from the corresponding untreated genotype.

Figure 2. Induction of p53-dependent apoptotic genes in p53 mice following irradiation in vivo

The head and neck region of p53+/+, p53+/−, and p53−/− mice was exposed to 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 hours post-irradiation. PUMA and Bax expression was analyzed as described in Figure 1. Induction of PUMA (A) and Bax (B) in p53 genotype mice. Significant differences (p≤0.05) were determined using an ANOVA followed by a pairwise Bonferroni test adjusting for the number of comparisons. (*) indicates significant difference from the corresponding untreated genotype.

Figure 3. Analysis of PERP following irradiation in vivo

The head and neck region of p53+/+, p53+/−, and p53−/− mice was exposed to 1, 2, or 5 Gy radiation and the parotid salivary glands were removed 4 hours post-irradiation. PERP expression was analyzed as described in Figure 1.

Figure 4. Reduced apoptosis in p53−/− mice following head and neck irradiation

In (A), the head and neck region of p53+/+, p53+/−, and p53−/− mice was exposed to 1, 2, 5, or 10 Gy radiation and the parotid salivary glands were removed 24 hours post-irradiation. The number of caspase-3 positive cells is graphed as a percentage of the total number of cells per field of view. In (B), four-week old female p53+/+, p53+/−, and p53−/− mice were exposed to 5 Gy radiation as described in (A) and parotid salivary glands were removed at 24, 48, 72, and 96 hours. Tissues were processed for activated caspase-3 immunohistochemistry as described in (A). The graph represents all data from three mice per group (exception: p53−/− mice were 2 mice per group). Significant differences (p≤0.05) were determined using an ANOVA followed by a pairwise Bonferroni test adjusting for the number of comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 4. Reduced apoptosis in p53−/− mice following head and neck irradiation

In (A), the head and neck region of p53+/+, p53+/−, and p53−/− mice was exposed to 1, 2, 5, or 10 Gy radiation and the parotid salivary glands were removed 24 hours post-irradiation. The number of caspase-3 positive cells is graphed as a percentage of the total number of cells per field of view. In (B), four-week old female p53+/+, p53+/−, and p53−/− mice were exposed to 5 Gy radiation as described in (A) and parotid salivary glands were removed at 24, 48, 72, and 96 hours. Tissues were processed for activated caspase-3 immunohistochemistry as described in (A). The graph represents all data from three mice per group (exception: p53−/− mice were 2 mice per group). Significant differences (p≤0.05) were determined using an ANOVA followed by a pairwise Bonferroni test adjusting for the number of comparisons. (*) indicates significant difference from the corresponding untreated genotype and (#) indicates significance between genotypes of the same treatment.

Figure 5. Acute salivary gland dysfunction 3 days after radiation exposure

p53+/+, p53+/−, and p53−/− mice were exposed to A) 2 Gy and B) 5 Gy radiation. Three days after exposure to radiation total saliva was collected following carbachol injection and graphed as μg/min. Graph A represents all data from at least eight mice/group and graph B represents all data from at least six mice/group. Significant differences (p≤0.05) were determined using an ANOVA followed by Tukey’s multiple comparison LSD. Treatment groups with the same letters are not significantly different from each other.

Figure 5. Acute salivary gland dysfunction 3 days after radiation exposure

p53+/+, p53+/−, and p53−/− mice were exposed to A) 2 Gy and B) 5 Gy radiation. Three days after exposure to radiation total saliva was collected following carbachol injection and graphed as μg/min. Graph A represents all data from at least eight mice/group and graph B represents all data from at least six mice/group. Significant differences (p≤0.05) were determined using an ANOVA followed by Tukey’s multiple comparison LSD. Treatment groups with the same letters are not significantly different from each other.

Figure 6. Chronic salivary gland dysfunction 30 days after radiation exposure

p53+/+, p53+/−, and p53−/− mice were exposed to A) 2 Gy and B) 5 Gy radiation. Thirty days after exposure to radiation total saliva was collected following carbachol injection and graphed as μg/min. Graph A represents all data from at least eight mice/group and graph B represents all data from at least six mice/group. Significant differences (p≤0.05) were determined using a One-Way ANOVA followed by Tukey’s multiple comparison LSD. Treatment groups with the same letters are not significantly different from each other.

Figure 6. Chronic salivary gland dysfunction 30 days after radiation exposure

p53+/+, p53+/−, and p53−/− mice were exposed to A) 2 Gy and B) 5 Gy radiation. Thirty days after exposure to radiation total saliva was collected following carbachol injection and graphed as μg/min. Graph A represents all data from at least eight mice/group and graph B represents all data from at least six mice/group. Significant differences (p≤0.05) were determined using a One-Way ANOVA followed by Tukey’s multiple comparison LSD. Treatment groups with the same letters are not significantly different from each other.