Systemic effects after ionizing radiation exposure: Genome-wide transcriptional analysis of mouse normal tissues exposed to (211)At, (131)I, or 4 MV photon beam

The radionuclides ¹³¹I and ²¹¹At are used or proposed for a variety of cancer treatments. Both radiohalogens exhibit considerable uptake in the thyroid gland and–to a lower degree–in various other tissues if they are unbound or released from a cancer-targeting agent. The resulting differential exposure throughout the body introduces two paradigms when studying effects in non-thyroid tissues in vivo: 1) low-dose effects of the organ, and 2) systemic effects due to a dominantly exposed regulatory organ. Normal tissue response is an important parameter in risk assessment in order to give the highest possible absorbed dose to malignant tumors while minimizing detrimental side effects in healthy tissue. The aim of this work was to increase fundamental knowledge of normal tissue responses to differential ionizing radiation exposure in vivo and to evaluate key findings regarding circadian rhythm and data convolution. Female BALB/c nude mice were used as a model system and the kidney cortex, kidney medulla, liver, lungs, spleen, and thyroid were studied using RNA microarray technology. Genome-wide transcriptional regulation and basically all analytical endpoints studied were tissue-specific. In various tissues, the <i>Angptl4</i>, <i>Per1</i> and <i>Per2</i>, and <i>Tsc22d3</i> genes may be potential biomarkers for ²¹¹At exposure (Papers I, II). In the thyroid, the <i>Klk1</i> gene-family may serve as biomarker candidates for ¹³¹I exposure (Paper V). Similarity in the extent of regulation irrespective of absorbed dose level generated a hypothesis on thyroid-dependent systemic effects in non-thyroid tissues, which was supported by gene signature and pathway analysis (Papers I, III). Results from partial body irradiation with 4 MV photon beams confirmed the hypothesis (Paper IV). Circadian rhythm affected the extent and quality of regulation in a tissue-specific manner, but key findings showed certain robustness to diurnal variation (Paper V). Deconvolution of microarray data increased detection rate of significantly differentially expressed transcripts in thyroid data, but also confirmed key results derived from convoluted data (Paper VI). In conclusion, low-dose exposure, systemic effects, and circadian rhythm have a pronounced impact on normal tissue response in vivo and should be considered for more accurate risk assessment in radionuclide therapy