Iodine-125-labeled DNA-Triplex-forming oligonucleotides reveal increased cyto- and genotoxic effectiveness compared to Phosphorus-32

PURPOSE

The efficacy of DNA-targeting radionuclide therapies might be strongly enhanced by employing short range particle-emitters. However, the gain of effectiveness is not yet well substantiated. We compared the Auger electron emitter I-125 to the ß^--emitter P-32 in terms of biological effectiveness per decay and radiation dose when located in the close proximity to DNA using DNA Triplex-forming oligonucleotides (TFO). The clonogenicity and the induction of DNA double-strand breaks (DSB) were investigated in SCL-II cells after exposure to P-32- or I-125-labeled TFO targeting the glyceraldehyde 3-phosphate dehydrogenase (GAPDH) gene and after external homogeneous exposure to gamma-rays as reference radiation.

MATERIALS AND METHODS

TFO were labeled with P-32 or I-125 using the primer extension method. Cell survival was analyzed by colony-forming assay and DNA damage was assessed by microscopic quantification of protein 53 binding protein 1 (53BP1) foci in SCL-II cells.

RESULTS

I-125-TFO induced a pronounced decrease of cell survival (D₃₇ at ∼360 accumulated decays per cell, equivalent to 1.22 Gy cell nucleus dose) and a significant increase of 53BP1 foci with increasing decays. The P-32-labeled TFO induced neither a strong decrease of cell survival nor an increase of 53BP1 foci up to ∼4000 accumulated decays per cell, equivalent to ∼1 Gy cell nucleus dose. The RBE for I-125-TFO was in the range of 3-4 for both biological endpoints.

CONCLUSIONS

I-125-TFO proved to be much more radiotoxic than P-32-TFO per decay and per unit dose although targeting the same sequence in the GAPDH gene. This might be well explained by the high number of low energy Auger electrons emitted by I-125 per decay, leading to a high ionization density in the immediate vicinity of the decay site, probably producing highly complex DNA lesions overcharging DNA repair mechanisms.

Fast online system for forecasting environmental impact during an incident

In an incident, the operational control staff needs a rapid general view of the radiological situation in the vicinity of the emission source which is suitable for decision support. Because the measuring team cannot provide sufficient measurements immediately, model calculations based on the available information about the course of the accident are additionally required. Such calculations do not only serve to create an overview of the current situation but can also provide prognoses for several hours. For this purpose, a software system for environmental impact forecasting was developed at Research Centre Jülich, which calculates time-integrated doses as well as ambient gamma dose rates and activity values. The latter can be directly compared with values measured during the incident and consequently can be used to assess the model assumptions.

Cellular response on Auger- and Beta-emitting nuclides: human embryonic stem cells (hESC) vs. keratinocytes

PURPOSE

We studied the response of human embryonic stem cells (hESC) to the β-emitter (131)I, which affects the entire cell and to the Auger electron emitter (125)I-deoxyuridine ((125)I-dU), primarily affecting the deoxyribonuleic acid (DNA). The effects were also studied in keratinocytes as a prototype for somatic cells.

METHODS

HESC (H1) and human keratinocytes (HaCaT, human) were exposed to (125)I-dU (5 × 10(-5) - 5 MBq/ml) and (131)I-iodide (5 × 10(-5) - 12.5 MBq/ml) and apoptosis was measured by DNA-fragmentation. Cell morphology was studied by light microscopy and electron microscopy. Transcriptional profiling was done on the Agilent oligonucleotide microarray platform.

RESULTS

Auger-process induced no apoptosis but a strong transcriptional response in hESC. In contrast, HaCaT cells showed a pronounced induction of apoptosis but only a moderate transcriptional response. Transcriptional response of hESC was similar after (125)I-dU and (131)I treatments, whereas HaCaT cells expressed a much more pronounced response to (125)I-dU than to (131)I. A striking radiation-induced down-regulation of pluripotency genes was observed in hESC whereas in keratinocytes the enriched gene annotations were related primarily to apoptosis, cell division and proliferation.

CONCLUSIONS

Human embryonic stem cells respond to ionizing radiation by (125)I-dU and (131)I in a different way compared to keratinocytes. Transcriptional response and gene expression appear to facilitate an escape from programmed cell death by striking a new path which probably leads to cell differentiation.

Charge build-up during decay of DNA-incorporated (123/125)I: consequences for labeled molecular structures

PURPOSE

To further elucidate the mechanisms behind the strong biological effectiveness of DNA-incorporated Auger electron emitters (123)I and (125)I, which are mostly attributed to the shower of low-energy electrons released during the decay. A second, frequently mentioned cause can be seen in the charges accumulated during the Auger cascade on the decaying nuclide and its subsequent intra-molecular redistribution leading to a Coulomb explosion.

METHODS

To assess the size of the charge and the dimensions of DNA damage thus determined, the first Auger cascade was simulated by Monte Carlo methods. The consequences of intra-molecular charge transfer in terms of structural molecular alterations were estimated by density functional theory (DFT) calculations and folding with the results of the Monte Carlo studies.

RESULTS

Charge distributions of (123)I and (125)I were found to be very similar with values between + 1 and + 15 and a mean value of + 6.4. The molecules could tolerate charges up to + 5 (base), + 2 (nucleoside) and + 7 (nucleotide) without being destroyed.

CONCLUSIONS

The strong molecular DNA damage after (123)I and (125)I decay depends very much on the size of the DNA molecule involved in the calculation. In general, not every decay can be expected to lead to a Coulomb explosion.

Computer evaluation of direct and indirect damage induced by free and DNA‐bound Iodine‐125 in the chromatin fibre

PURPOSE

When Iodine-125 decays within chromatin, several in vivo experiments have shown that the radiobiological effects are caused mainly by indirect mechanisms and that more than one DNA double strand break (DSB) is produced per decay. We present calculations to evaluate the contribution of direct and indirect effects of radiation tracks to produce DNA damage induced by bound and free I-125 in a model of chromatin DNA.

MATERIALS AND METHODS

A solenoid model of chromatin with 18 nucleosomal elements placed in bulk water (more than 600,000 atoms) is used where the initial I-125 decay takes place. All physical and chemical events initiated by Auger and X-rays were taken into account. The yields of single strand breaks (SSB) and DSB were derived using direct effects on DNA and indirect reactions of all radical species generated in the radiolysis of the bulk water.

RESULTS

The distribution of damage complexity for free and DNA-bound I-125 is presented. We obtained more than 1.3 DSB per decay, with nearly equal contributions from direct and indirect effects. However, for the most complex type of damage, located at the decay site, the direct effect is about 70% of the total number. To show the protective effect of histones, simulations were carried out with and without the presence of histones.

Is Coulomb explosion a damaging mechanism for125IUdR?

PURPOSE

To test the integrity of the thymine molecule that experiences an increasing number of charges due to the loss of Auger electrons emitted by the decay of incorporated 125I. Besides the radiation action of these electrons, Coulomb explosion is suspected to be an additional mechanism responsible for the strong radiotoxic effect of decaying DNA-incorporated 125I. The two-step decay process initiates a first Auger cascade within 10(-16) to 10(-14) s resulting in the release of about 7 electrons on average and a corresponding large positive charge on the 125Te daughter atom. Being part of iododeoxyuridine (125IUdR), the analogue of the DNA base thymine, the base is suddenly confronted with this charge. Experimentally, the situation was investigated with small molecules (CH3(125)I and C2H5(125)I) resulting in ion fragmentation in agreement with a Coulomb explosion model (Carlson and White, 1963, 1966).

MATERIALS AND METHODS

Semi-empirical quantum mechanical calculations on the Parametric Method 3 (PM3) level (Stewart, 1989a, 1989b) were performed and geometry optimisation was applied for the identification of stable molecule conformations. Subsequently, semiempirical molecular dynamics simulations allowed changes in the conformations to be studied as a function of time.

RESULTS

First results show that there is no stable molecular configuration with a total charge of > or = +5e. PM3 calculations will not converge for such a charge located at the 125I/125Te position. This finding is supported by total energy considerations, which begin to favour a system of isolated atoms versus molecular bound atoms when the molecular charge is greater than +4e. The distribution of the partial charges indicates that most of the charge will remain on the tellurium atom with slight increases of charge at the other molecular partners within 125IUdR. Moreover, the molecular dynamics simulations reveal a breaking of chemical bonds between those atoms with the strongest charge increase.

CONCLUSIONS

Coulomb explosion must be taken into account as a possible damaging mechanism following the decay of DNA-incorporated Auger electron emitters. Lobachevsky and Martin (2000) have identified the same mechanism to be responsible for part of strand breakage in oligo-deoxynucleotides. To elucidate a possible link between both damage patterns the molecular mechanics simulations have to be extended to larger parts of the DNA molecule.

Computer simulation of strand break yields in plasmid pBR322: DNA damage following 125I decay

This paper presents results of (125)I effects on plasmid pBR322 in aqueous solution, simulating the complete transport of Auger and X rays up to the chemical phase. In addition to new sampling algorithms, new electronic cross sections are included. Simulations were carried out both with (125)I, bound to plasmid, or free, in its vicinity. The influence of the hydroxyl radical scavenger dimethyl sulfoxyde (DMSO) has also been tested, underlying that, in naked DNA, double strand breaks (caused by the decay of bound (125)I) are mainly due to direct hits. The calculated yields of relaxation events (RE) and linearization events (LE) show good agreement with experimental ones: when (125)I is bound to the plasmid pBR322, 0.16 RE and 0.83 LE per decay (without DMSO) are then observed. Then, when 2 mol DMSO is added, RE and LE probabilities become 0.22 and 0.76. The very light differences with those from literature could arise from experimental conditions.

Estimation of a radiation weighting factor for 99mTc

Decaying (99m)Tc does not only emit a gamma ray (140.5 keV), but also low-energy Auger and conversion electrons. These electrons cause a serious problem in the determination of a radiation weighting factor for (99m)Tc due to their extremely short range in tissue. Therefore, for comparison ultrasoft X rays are used here, which deposit their energy mainly via the photoeffect thus also initiating low-energy photoelectrons. Monte Carlo computer codes provided electron emission spectra of (99m)Tc and subsequent track structure calculations simulated the induction of DNA damage of different degrees of complexity. For the modelling of ultrasoft X rays carbon K photons with an energy of 270 eV were selected, for which experimental results are available from the literature. On average, four electrons were found to be emitted per (99m)Tc decay. Simulation of DNA damage revealed a nearly identical spectrum of primary strand breaks for (99m)Tc and C-K radiation. On this basis, a total radiation weighting factor of 1.2 was evaluated for (99m)Tc.

A computer-generated supercoiled model of the pUC19 plasmid

DNA models have become a powerful tool in the simulation of radiation-induced molecular damage. Here, a computer code was developed which calculates the coordinates of individual atoms in supercoiled plasmid DNA. In this prototype study, the known base-pair sequence of the pUC19 plasmid has been utilized. The model was built in a three-step process. Firstly, a Monte Carlo simulation was performed to shape a segment chain skeleton. Checks on elastic energy, distance and unknotting were applied. The temperature was considered in two different ways: (1) it was kept constant at 293 K and (2) it was gradually reduced from 350 K to less than 10 K. Secondly, a special smoothing procedure was introduced here to remove the edges from the segment chain without changing the total curve length while avoiding the production of overshooting arcs. Finally, the base pair sequence was placed along the smoothed segment chain and the positions of all the atoms were calculated. As a first result, a few examples of the supercoiled plasmid models will be presented, demonstrating the strong influence of appropriate control of the system temperature.

Computer simulation of 57Fe bleomycin auger effects in DNA

The antibiotic bleomycin binds to the DNA and induces double strand breaks (DSBs). To increase the cleavages. 57Fe is used to form a complex suitable for Mössbauer effect. The de-excitation of the resonant excited 57Fe nucleus releases Auger electrons and X rays. The goal of this work is to evaluate the increase in yield of DSBs due to the 57Fe, using Monte Carlo simulation methods. Particles spectra and the yields of single strand breaks (SSBs) and DSBs were calculated by considering direct events on DNA and reaction of all radical species generated in the radiolysis of its environment. The Auger spectrum shows a large number of electrons with energies below 100 eV, mainly responsible for direct damage, while another group around 600-700 eV is responsible for indirect damage effects. Bleomycin receives about one fourth of the energy deposited in DNA and an average of 0.65 DSB per de-excitation is observed.

Ratio of complex double strand break damage induced by 125IUdR and 123IUdR correlates with experimental in vitro cell killing effectiveness

The overall cellular damage induced by ionising radiation is determined by the number and spatial distribution of initial ionisations and excitations within the critical volume. This paper focuses on the physical and chemical phase of the radiation action chain following the decay of DNA-bound 123I and 125I. Monte Carlo simulations of these nuclides' decay provide electron emission spectra which are used as input data for track structure calculations. In combination with DNA models, these calculations allow the specific radiation source to be characterised in terms of DNA strand break patterns. The distribution of these patterns indicates that 125I produces much more severe breaks than 123I. The ratio of complex DSBs induced by both iodine isotopes correlates with the differences in cell killing effectiveness reported from in vitro survival experiments.

Track structures and dose distributions from decays of (131)I and (125)I in and around water spheres simulating micrometastases of differentiated thyroid cancer

The disintegration of the radionuclides (131)I and (125)I and the subsequent charged-particle tracks left behind in water (as a model substance for a biological cell) are simulated by the Monte Carlo track structure simulation code PARTRAC, using new inelastic electron scattering cross sections for condensed water. Every photon and electron emitted was followed in detail, event by event, down to 10 eV. From the spatial information on the track structures, absorbed dose distributions per (131)I and (125)I decay were calculated in and around water spheres simulating micrometastases as well as in the tissue surrounding such metastases. These radionuclides were assumed to be distributed uniformly inside spheres of different diameters (0.01, 0.03, 0.1, 0.3, 1.0 and 3.0 mm). The respective electron degradation spectra, the nearest-neighbor distance distributions between inelastic events, and the distance distributions for all activations for both iodine radionuclides were calculated. The absorbed fractions of the initial electron energies, absorbed doses and energy depositions, and single-event distributions, F(1)(epsilon), inside the six water spheres described above and in the surrounding tissue were also calculated. The absorbed doses per decay inside the six water spheres, i.e., the calculated S values (listed from 0.01 to 3.0 mm), were 6.8 x 10(-4), 7.2 x 10(-5), 5.5 x 10(-6), 4.9 x 10(-7), 3.1 x 10(-8) and 1.8 x 10(-9) Gy Bq(-1) s(-1) for (131)I, and 3.4 x 10(-3), 1.7 x 10(-4), 5.1 x 10(-6), 2.0 x 10(-7), 5.6 x 10(-9) and 2.2 x 10(-10) Gy Bq(-1) s(-1) for (125)I. It is concluded that, in the treatment of thyroid cancer, the geometrical track structure properties of (125)I might be superior to those of (131)I in micrometastases with diameters less than 0.1 mm; however, in this medical context, many other factors also have to be considered.

Auger electron spectra--the basic data for understanding the Auger effect

Understanding the strong radiotoxicity of DNA-incorporated Auger electron-emitting nuclides requires a detailed knowledge of the nuclide's emission spectrum. A Monte Carlo computer code was previously developed to simulate Auger cascades and to provide electron spectra of 125I. To utilize experimental data for a direct validation of these simulations, the code has been adapted for cascades in xenon, which is adjacent to iodine in the table of elements. Only minor modifications of the code were necessary to obtain a very good agreement with the experimental findings. The role of shake-off electrons and the need for energy considerations during the cascades could be demonstrated. A previously published electron spectrum of 125I was recalculated and detailed results are presented here. Furthermore, to consider implications from a molecular binding of the Auger emitter, for the first time semi-empirical quantum mechanical calculations for an iodine-labelled thymine molecule were performed showing that even in the condensed phase a Coulomb explosion cannot be excluded a priori.

Modelling of initial events and chemical behaviour of species induced in DNA units by Auger electrons from 125I, 123I and carbon

Auger electron spectra for 123I and 125I generated by Monte Carlo calculation and Auger electrons emitted from carbon after photoelectric effect on its K-shell as well as two DNA models (linear plasmid and nucleosome model) based on x-ray diffraction experiments have been used to simulate the behaviour of all species and radicals created during the physical and the chemical phase of the Auger's transport. By introducing appropriate assumptions for the induction of strand breaks the number of these breaks can also be determined and correlated to experimentally found numbers of lethal events. Efficiency differences between the iodine nuclides themselves and in comparison with the rather monoenergetic Auger electrons from carbon are shown with regard to the direct and indirect effects on the two DNA models. The characteristic products in the physical, chemical and biochemical phase are compared with corresponding results from the literature for low-LET radiation.

A nucleosome model for the simulation of DNA strand break experiments

Using a set of Monte Carlo simulation models, track structures of 125I Auger electrons generated in liquid water are superimposed on a nucleosome DNA model able to precisely localize energy deposition events on sub-molecular units of the DNA strands. After scoring direct hits taking place during the physical phase (at about 10(-15) s) the radiation chemistry of the whole system is simulated between 10(-12) and 10(-8) s, taking into account all reactions between water radio-chemical species, radicals, sub-molecular units of DNA (Ribose, Adenine, Thymine, Guanine, and Cytosine), and scavengers like Tris or Formate ions. The model's possibility to distinguish between direct and indirect hits has been utilized to introduce different assumptions for strand break induction by both hit modes. The number of SSB and DSB as well as their local distribution will be given and compared with experimental and theoretical results from the literature.

Low-energy electrons inside active DNA models: a tool to elucidate the radiation action mechanisms

To postulate radiation action mechanisms and to test them by Monte Carlo simulation, a complex computer model was developed consisting of major components for the generation of a radiation spectrum, biomolecular structures, and electron track structures in liquid water. As the radiation source 125I is employed here; it is an excellent test radiation due to its exactly localized position in the DNA molecule and high biological toxicity as a consequence of the emission of short-ranging Auger electrons. A linear DNA plasmid model (Pomplun 1991) which can actively respond to radical attack (Terrissol and Pomplun 1994) has been modified into a nucleosome model representing the double-helix of DNA with 146 basepairs and more than 9000 atoms surrounding the histones. The introduction of this new target structure allows a more realistic simulation of cellular conditions. Using the model's decay accumulation aspect, the situation of many break and survival experiments can be approximated and the influence of several cellular parameters tested. As a first step, a correlation between the size of energy depositions and strand-break patterns was sought.

A new DNA target model for track structure calculations and its first application to I-125 Auger electrons

A DNA target model has been developed, based on the geometrical co-ordinates of individual atoms. This model is used to analyse DNA damage produced by Auger electron tracks from the decay of 125I. The high resolution of this target model enables the distinction between direct and indirect electron hits, i.e. hits inside the atomic volumes of the DNA molecule and those hitting the water molecules assumed in the space between the atomic volumes. Three types of calculations have been performed: (1) the evaluation of the energy deposition in the surroundings of the decaying 125I nuclide demonstrating different fractions of direct to indirect hits at different parts of the DNA molecule (phosphate/sugar strand or bases), (2) a detailed energy deposition pattern in the radiolabelled base, indicating that this most burdened molecule is not necessarily destroyed by direct hits, and (3) a calculation of single- and double- strand breaks by using different threshold values for effective direct and indirect hits, resulting in a good correlation with experimental data on strand break efficiency.

A Monte Carlo simulation of Auger cascades

The energy imparted to biological tissue after the decay of incorporated Auger emitters stems from two sources: (a) energy deposition by the Auger and Coster-Kronig electrons and (b) the charge potential which remains on the multiple ionized atom after the end of the cascade. For the numerical assessment of both the kinetic energy of the released electrons and the charge potential, a new and--for purposes of microdosimetry--precise method is presented. Based on relativistic Dirac-Fock calculations and a rigorous bookkeeping, this method provides a perfect energy balance of the considered atomic system when applied to Monte Carlo simulations of Auger cascades. By comparing the results for charge distribution for krypton and iodine with experimental data and the electron spectrum of 125I with theoretical data, it can be shown that the approach followed in this work is reasonable and appropriate for the determination of the energy deposited by incorporated Auger emitters in small volumes of condensed matter. The total energy deposited by 125I in a volume of 20-nm diameter is 2.03 keV which is made up by multiple ionization (1.07 keV) and energy deposition by the emitted Auger electrons (0.96 keV).

Some consequences of the Auger effect: fluorescence yield, charge potential, and energy imparted

The potential energy produced by the Auger cascade due to the charging of atoms is evaluated and incorporated into conventional treatment of energy deposition. A straightforward method for calculating this energy is presented. For the photoelectric interaction the potential energy is shown to be at least as important as L-shell fluorescence in calculating the electron kerma. For radioactive decay by electron capture or internal conversion, it is shown that, for small (less than 100 nm) targets containing the decay, the atomic charging can be the dominant contribution to the total energy deposited in the target.