Effects of Radiation Quality and Oxygen on Clustered DNA Lesions and Cell Death

Robert D. Stewarta,†, Victor K. Yub,c, Alexandros G. Georgakilasd, Constantinos Koumenise, Joo Han Parkc, David J. Carlsonb
Radiat. Res. 176, 587-602 (2011). Download PDF

a Department of Radiation Oncology, University of Washington, Seattle, WA 98195-6043
 b Department of Therapeutic Radiology, Yale University School of Medicine, New Haven, CT 06520-8040
 c School of Health Sciences, Purdue University, West Lafayette, IN 47907-2051
d Department of Biology, Thomas Harriot College of Arts and Sciences, East Carolina University, Greenville, NC 27858-4353, USA
 e University of Pennsylvania School of Medicine, Department of Radiation Oncology, Philadelphia, PA 19104-6072 

* VKY and RDS contributed equally to the project. AGG and CK are responsible for measurements of DSB, Fpg and Endo III cluster yields in HeLa cells. VKY, RDS and DJC contributed to the collection, analysis and interpretation of data. JHP performed MCNP simulations of secondary electron energy spectra for photons. RDS is responsible for the overall conception and design of the project.  All authors contributed to preparation of the manuscript and have approved of the final manuscript.

Abstract

Radiation quality and cellular oxygen concentration have a substantial impact on DNA damage, reproductive cell death and, ultimately, the potential efficacy of radiation therapy for the treatment of cancer. To better understand and quantify the effects of radiation quality and oxygen on the induction of clustered DNA lesions, we have now extended the Monte Carlo Damage Simulation (MCDS) to account for reductions in the initial lesion yield arising from enhanced chemical repair of DNA radicals under hypoxic conditions. The kinetic energy range and types of particles considered in the MCDS have also been expanded to include charged particles up to and including 56Fe ions.  The induction of individual and clustered DNA lesions for arbitrary mixtures of different types of radiation can now be directly simulated. For low linear energy transfer (LET) radiations, cells irradiated under normoxic conditions sustain about 2.9 times as many double strand breaks (DSB) as cells irradiated under anoxic conditions. New experiments performed by us demonstrate similar trends in the yields of non-DSB (Fpg and Endo III) clusters in HeLa cells irradiated by g-rays under aerobic and hypoxic conditions.  The good agreement among measured and predicted DSB, Fpg and Endo III cluster yields suggests that, for the first time, it may be possible to determine nucleotide-level maps of the multitude of different types of clustered DNA lesions formed in cells under reduced oxygen conditions.

 As particle LET increases, the MCDS predicts that the ratio of DSB formed under normoxic to hypoxic conditions by the same type of radiation decreases monotonically towards unity.  However, the relative biological effectiveness (RBE) of higher LET radiations compared to 60Co g-rays (0.24 keV/mm) tends to increase with decreasing oxygen concentration.  The predicted RBE for DSB induction of a 1 MeV proton (26.9 keV/mm) relative to 60Co g-rays increases from 1.9 to 2.3 as oxygen concentration decreases from 100% to 0%.  For a 12 MeV 12C ion (681 keV/mm), the predicted RBE for DSB induction increases from 3.4 (100% O2) to 9.8 (0% O2).  Estimates of linear-quadratic (LQ) cell survival model parameters (a and b) are closely correlated to the Monte Carlo predicted trends in DSB induction for a wide range of particle types, energies and oxygen concentrations.  The analysis suggests a is, as a first approximation, proportional to the initial number of DSB per cell, and b is proportional to the square of the initial number of DSB per cell.  Although the reported studies provide some evidence supporting the hypothesis that DSB are a biologically critical form of clustered DNA lesion, the induction of Fpg and Endo III clusters in HeLa cells irradiated by g-rays exhibit similar trends with oxygen concentration.  Other types of non-DSB cluster may still play an important role in reproductive cell death.  The MCDS captures many of the essential trends in the formation of clustered DNA lesions by ionizing radiation and provides useful information to probe the multi-scale effects and interactions of ionizing radiation in cells and tissues.  Information from Monte Carlo simulations of cluster induction may also prove useful for efforts to better exploit radiation quality and reduce the impact of tumor hypoxia in proton and carbon ion radiation therapy.

Acknowledgements

The authors thank Dr. Vladimir A. Semenenko for developing the original MCDS algorithm and software and for providing useful feedback and suggestions on the current manuscript. The authors also thank Mr. Anshuman Panda for implementing an early version of the model for chemical repair and oxygen fixation. Work supported in part by American Cancer Society grant IRG-58-012-52 (DJC) and in part by a Research Creative Activity Grant from the East Carolina University (ECU) Biology Department (AGG).

Last updated: September 14, 2015
Contact Webmaster