# Full Text: NuclearStructure

> Extracted from `2016_NuclearStructure.pdf`

---

## Page 1

Full Terms & Conditions of access and use can be found at
http://www.tandfonline.com/action/journalInformation?journalCode=irab20
Download by: [University of California, San Diego]
Date: 14 March 2016, At: 06:07
International Journal of Radiation Biology
ISSN: 0955-3002 (Print) 1362-3095 (Online) Journal homepage: http://www.tandfonline.com/loi/irab20
Influence of nuclear structure on the formation of
radiation-induced lethal lesions
Daniel A. Friedman, Lauren Tait & Andrew T. M. Vaughan
To cite this article: Daniel A. Friedman, Lauren Tait & Andrew T. M. Vaughan (2016): Influence
of nuclear structure on the formation of radiation-induced lethal lesions, International Journal
of Radiation Biology, DOI: 10.3109/09553002.2016.1144941
To link to this article:  http://dx.doi.org/10.3109/09553002.2016.1144941
Published online: 26 Feb 2016.
Submit your article to this journal 
Article views: 38
View related articles 
View Crossmark data

## Page 2

REVIEW
Influence of nuclear structure on the formation of radiation-induced lethal
lesions
Daniel A. Friedmana, Lauren Taitb and Andrew T. M. Vaughanb
aDepartment of Biology, Gilbert Hall, Stanford University, California; bDepartment of Radiation Oncology, University of California, Davis,
Sacramento, California, USA
ABSTRACT
Purpose The rejoining of fragmented nuclear DNA caused by ionizing radiation may lead to lethal
chromosome rearrangements, such as rings or dicentrics. The clinically useful linear quadratic relation-
ship between dose and cell survival has been interpreted as the generation of lethal lesions secondary
to damage occurring in two separate chromosomes simultaneously (a component), or as potentially
repairable separate events (b component). Here, the generation of such lesions is discussed, synthesiz-
ing existing knowledge with new insights gleaned from spatial proximity data made possible by high-
throughput sequencing of chromosome conformation capture experiments. Over a range of several
Mbp, the linear DNA strand is organized as a fractal globule generating multiple sites of contact that
may facilitate deletions or inversions if the points of contact are damaged. On a larger scale, transcrip-
tionally active euchromatin occupies a physically identifiable space separate from inactive areas and is
preferentially susceptible to free radical attack after irradiation. Specific transcriptional programs link
genomic locations within that space, potentially enhancing their interaction if subject to simultaneous
fragmentation by a single radiation event.
Conclusions High throughput spatial analysis of the factors that control chromosome proximity has the
potential to better describe the formation of the lethal chromosome aberrations that kill irradiated cells.
ARTICLE HISTORY
Received 4 February 2015
Revised 9 December 2015
Accepted 13 January 2016
KEYWORDS
Cellular radiobiology;
chromosome abberations;
models of cell killing;
double-strand breaks;
molecular radiobiology;
microdosimetry
Introduction
Over the past five decades, developments in treatment deliv-
ery, computerization and imaging have enabled the practice
of
radiation
oncology
to
make
substantial
advances.
Megavoltage accelerators have provided improved depth
dose profiles while subsequent onboard imaging technology
has helped spare more normal tissue from damage (Jaffray
2012). Throughout the substantial technological advances
that have significantly improved treatment outcomes, a com-
prehensive description of the process whereby radiation gen-
erates lethal lesions is lacking. The biological focus has rightly
been the response of the cell to radiation-induced DNA dou-
ble-strand breaks, the key lethal radiation event, their signal-
ing to regulatory pathways and the mechanism(s) of their
repair (Jackson and Bartek 2009, Thompson 2012, Price and
D’Andrea 2013, Kavanagh et al. 2013). However technical limi-
tations in detecting the location of such breaks, have limited
studies on their potential to form a lethal lesion that restricts
mitosis, such as a dicentric, ring structure or anaphase bridge
(Costes et al. 2007). At least part of the reason for this dis-
crepancy is the architectural complexity of the genome. At its
simplest, two DNA double-strand breaks generated by irradi-
ation within the same or separate chromosomes need to be
ligated together producing an aberration rather than be indi-
vidually repaired. The ability to generate such lesions is there-
fore determined by the spatial distribution of radiation
energy within the complex structure of the genome itself.
It is clear the latter aspect has lagged the development of
sophisticated models of radiation deposition that describe
radiation
interactions
(Goodhead
2006).
However
recent
advances in the field(s) of genomic analysis have provided a
more detailed spatial description of the genome that provide
a more robust platform to map the generation of radiation
induced rearrangements (Lieberman-Aiden et al. 2009).
In addition to the spatial orientation of radiation-induced
lesions within the genome, their generation is also con-
strained by time dependent factors that exert their effects
over a very large range (Figure 1). Thus the earliest detectable
events within the genome, the physical photon or particle
interaction with biological material, will occur in the range of
femtoseconds. Subsequently, activated chemical species, pri-
marily free radicals, are generated and react over a range of
nano to microseconds with all cellular constituents, including
DNA. Actual lesion processing, through either repair or aber-
rant ligation, will be resolved over a period of hours.
Therefore the introduction of lethal lesions by irradiation is
impacted both by the complex spatial orientation of its gen-
omic target and the sequential development of events that
operate over an extremely broad time scale.
Radiation interactions in human cells
There is a long and rich history modelling the generation of
DNA breaks and lethal lesions following irradiation of human
CONTACT Dr Andrew Vaughan, PhD
atvaughan@ucdavis.edu
University of California, Davis, Radiation Oncology, 4501 X Street, Sacramento, CA 95817, USA
 2016 Taylor & Francis
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY, 2016
http://dx.doi.org/10.3109/09553002.2016.1144941
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 3

cells. Intrinsic to all of these is the discrete distribution of ion-
ization events throughout the cell. Of such models, the
‘Target Theory’ of Lea was one of the first to propose the
presence of multiple nuclear ‘targets’ that when ‘hit’ or inacti-
vated lead to the death of the cell (Lea 1955). In its ‘single
hit – multiple target’ formulation of Equation (1), S is the
fraction surviving dose D, D0 describes the slope of the expo-
nential portion of the survival curve and the extrapolation
number, n, the width of the shoulder region.
SðDÞ ¼ 1  ð1  eD=D0Þn
ð1Þ
In this form the probability of inactivating any one target
is the same, cell death will follow when a specific number of
targets are inactivated, the number predicted by the size of
the extrapolation number, n. Other variants such as the ‘mul-
tiple hits – multiple target’ model have been described that
increase the overall model complexity but still use the same
general approach (Nomiya 2013). The overall strategy was
however criticized by many in that no clear biological entity
corresponded with the proposed targets; however, the key
emphasis on sub-nuclear volumes as mediators of nuclear
change was, and still is, a valid approach in consideration of
the discontinuous distribution of energy in irradiated systems.
Subsequently, two additional models, the ‘Molecular the-
ory’ of Chadwick and Leenhouts and the ‘Theory of Dual
Radiation Action’ described by Kellerer and Rossi have been
described that, though starting from different sets of assump-
tions, each arrives at a similar description of cell kill, that of a
linear
quadratic
relationship
between
survival
and
dose
(Chadwick and Leenhouts 1973, Kellerer and Rossi 2012).
Here, both in vitro cell killing and in vivo responses can be
described as an effect ‘E’, dependent on the dose ‘d’ deliv-
ered over ‘n’ fractions, at least up to moderate doses
(Equation 2).
E ¼ nðad þ bd2Þ
ð2Þ
In
the
‘Molecular
theory’
model
of
Chadwick
and
Leenhouts, the significance of the DNA double-strand break
and its repair were used to derive a linear quadratic (LQ)
association
with
survival
after
radiation
(Chadwick
and
Leenhouts 1973). However this model incorporated a role for
the recombination of proximal DNA single-strand breaks as a
source for generating DNA double-strand breaks, an idea that
has not gained wide acceptance due to the likely rarity of
two such independent events occurring in close proximity.
Perhaps the most significant treatment in terms of its long-
term impact to the field is the ‘Theory of Dual Radiation
Action’
described
by
Kellerer
and
Rossi
(2012).
In
this
approach they incorporated the relatively new technology of
microdosimetry, utilizing large detectors filled with a gas that
simulated the density of very small sub-nuclear volumes.
They proposed that irradiation generated a population of
sub-lesions that had the potential to interact to cause a lethal
lesion, if they were in sufficiently close proximity. Such lethal
lesions clearly include the generation of chromosome rear-
rangements, such as dicentrics and more complex aberra-
tions, which interfere with cell division (Ballarini 2010, Hall
and Giaccia 2012). The interaction distances calculated were
of the order of a micrometer, a finding that quickly ran coun-
ter to experimental observation following the experiments of
Goodhead et al. (1979). In these later series of experiments,
using very low energy irradiation from ultrasoft aluminum or
carbon K edge X-rays, it was shown that the interaction dis-
tances of such sub-lesions was much closer than the microm-
eter proposed in the model of Kellerer and Rossi (Goodhead
et al. 1979). Nevertheless the LQ approach has proved to be
a robust model of X- or gamma-ray-induced cell killing that
has provided a useful and biologically grounded framework
to view experimental data. For the purpose of this review the
role of such modelling strategies is clear, they are clearly rela-
tively simple, and therefore flexible tools to predict the likely
survival of cells post irradiation. Viewed in this light they have
provided useful service, especially in the clinic, where the lin-
ear quadratic formalism has provided key information on the
relative potency of various fractionation strategies for treat-
ment (Fowler 2010). However, none are likely to offer the final
word in describing the effects of ionizing irradiation in com-
plex biological systems. As an example, the recent utilization
of a few large fractions to treat a range of cancers has been
modelled using a combination of both the linear quadratic
and target theory approaches to better describe anomalies of
the LQ formulae in high dose regions (Park et al. 2008).
Microdosimetry at the biological interface
The models discussed above emphasize a single outcome of
irradiation, cell survival vs. death, and include relatively simple
assumptions. In contrast, the study of microdosimetry, the
physical description of ionization energy deposition at the
level of discrete photon/electron interactions, has benefited
greatly from both the development of tissue equivalent pro-
portional counters and Monte Carlo-based analyses of track
structure (Goodhead 2006, Wiklund et al. 2011). These tools
have provided detailed and three-dimensional descriptions of
radiation interactions with targets the size, composition and
density of cells. The energy from incident photons may be
dissipated as localized clusters of ionizations and generation
of free radicals that have the potential to initiate both chem-
ical changes in the cell or physical disruptions, such as DNA
strand breaks. The radiation chemistry of such interactions
has been extensively reviewed and will not be repeated in
detail here (O’Neill and Wardman 2009). Energetic photons,
such as those used in radiation therapy, produce DNA dam-
age directly and indirectly. Directly, high-energy photons
can
break
the
phosphodiester
backbone
of
DNA.
DNA breaks
OH•, H•
Damage signaling
10−12
10−9
10−6
10−3
10−1
103
seconds
H20 radius •OH diffusion
10−12
10−9
10−8
10−7
10−6
10−5
meters
DNA width
DNA ends move
H2, OH-,H2O2
Ligation of DNA ends
Physics and Chemistry
Biology
Figure 1. Scale of radiation-induced events in both time and space. Direct DNA
damage and free radical mediated events such as those produced by the
hydroxyl and hydrogen radicals (OH•, H•) occur over a time scale that is orders of
magnitude faster and within a dimension that is also much smaller than the
recognizable lesions they produce.
2
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 4

Indirectly, high-energy photons can produce free radicals
through the homolytic cleavage of covalent bonds in nuclear
molecules, such as water. Free radicals then may react with
molecules in spatial proximity, which can include the DNA
phosphodiester
backbone.
Either
process
may
generate
breaks in one or both strands of DNA (Figure 1). The transfer
of photon energy to the target molecules is complete within
picoseconds and the free radicals execute chemical change
within microseconds. Thus the resolution of the initial phys-
ical and chemical events following irradiation is complete
within a time frame where the DNA organization may be con-
sidered a
static
entity,
a ‘snapshot’
of fixed geometry.
Subsequently, the resolution of damaged DNA proceeds at a
many magnitudes slower pace, over a time frame of hours,
where DNA fragments will be able to move in respect to
both each other and the rest of the genome.
Substantial amounts of data, generated over many deca-
des, have highlighted the DNA double-strand break as the
lesion most closely linked to cell death (Iliakis 1991, Olive
1998, Xu and Price 2011). For a double-strand break, each
end may physically separate prior to attempts at repair by
either
the
Non
Homologous
End
Joining
(NHEJ),
Microhomology Mediated End Joining (MMEJ) or Homologous
recombination
(HR)
pathways
(McVey
and
Lee
2008,
Chapman et al. 2012, Kavanagh et al. 2013, Schipler and
Iliakis 2013). Such a separation of free DNA ends may compli-
cate repair and lead to the generation of a lethal lesion, such
as a dicentric, as the free ends encounter alternate breaks.
Thus both the local environment of the lesion(s) and activity
of appropriate repair pathways may either promote or sup-
press such errors in DNA rejoining (Pfeiffer et al. 2004).
Perhaps the most convincing support for the key role of such
breaks in mediating lethality is the documented increase in
radiation sensitivity for cells that lack key components of the
DNA double-strand break repair pathways (Adachi et al. 2001,
Woodbine et al. 2014).
Most microdosimetry-based studies on radiation-induced
lethality have been limited by the lack of a comparably
detailed three-dimensional description of DNA organization,
its presumed target. Despite this drawback, studies have been
undertaken using such data that is available, including a
Monte Carlo simulation of radiation events matched with the
generation of DNA double-strand breaks and subsequent
dicentric formation (Edwards et al. 1996). The model success-
fully recapitulated the LQ relationship between dose and the
generation of dicentric lesions discussed above, though utiliz-
ing only a random distribution of DNA breaks within a regular
nuclear volume as its biological input. Others have included
the observation that nuclear DNA is packed into constrained,
supercoiled DNA loops and have modelled this level of organ-
ization as the targets for single or complex distributions of
lesions (Cook and Brazell 1976, Yokota et al. 1995, Khodarev
et al. 1997, Herr et al. 2014). Such studies clearly represent an
advance over the assumption of a random distribution of
DNA breaks within the genome, and they may readily recap-
itulate the admittedly simple LQ relationship between dose
and cell survival. However the low-complexity assumptions of
biological distributions that are employed prohibit any pro-
spective assessments of the role of nuclear architecture on
survival. Perhaps the most comprehensive series of studies to
date are represented by those incorporating the PARTRAC
(PARticle TRACk) code, combining Monte Carlo analysis of ion-
izing radiation track structure with multi-level modeling of
DNA
organization
(Friedland
et
al.
2011,
Friedland
and
Kundrat 2013). In this application, individual structures such
as nucleosomes, looped DNA, hetero and euchromatin are
modeled as defined structural entities, in addition to the
action of repair processes targeting the key DNA double-
strand break. Such a system has potential to help in the study
of radiation-induced lesions in the context of classically
organized nuclear DNA, though in its current version it does
not accurately predict the number of cytotoxic dicentrics that
are observed experimentally.
In this review we will emphasize the two extremes of radi-
ation-induced lethality, the nature and location of the initial
breaks, and the impact of genomic organization on the for-
mation and subsequent biological effect of chromosome
aberrations that are generated.
The lethal lesion
All the models discussed above are compatible with the
lethal lesion introduced by irradiation being the result of the
inappropriate fusion of two DNA double-strand breaks, gener-
ating toxic rearrangements such as a dicentric. Ionizing radi-
ation cleaves DNA by physical and/or chemical attack, leaving
DNA ends that contain a range of chemical moieties that
need to be processed to ‘clean’ 50
phosphates and 30
hydroxyl termini to facilitate ligation (Obe et al. 2010, Schipler
and Iliakis 2013, Averbeck et al. 2014). It is clear this process
is highly efficient in normal cells where the majority of dou-
ble-strand breaks are successfully repaired. The subsequent
inappropriate ligation of two such breaks on different chro-
mosomes can however generate an aberration with two cen-
tromeres (a dicentric chromosome) that will stop cell division
by physically restricting the partition of daughter cells at
mitosis as each linked centromere is pulled to opposite poles
of the cell. Specifically, the linear and quadratic terms within
the LQ equation are commonly assumed to represent DNA
fragmentation events that are generated either simultan-
eously, such as within a specific particle track (a), or from dis-
crete, separate, ionization events separated in space and time
(b) (Brown and Attardi 2005, Vakifahmetoglu et al. 2008,
Ballarini 2010, Hall and Giaccia 2012). Such an understanding
is consistent with all the modeling procedures discussed
above. In addition, the relevance of the LQ model in particu-
lar has been confirmed by decades of its practical application
in a clinical setting, where it has proven capable of predictive
power in illustrating the potency of a wide range of treat-
ment strategies. Here, the biologically effective dose (BED), of
treatment schedules of varying fraction size (d) can be deter-
mined in relation to variations in the dose response curve as
represented by the ratio of a and b (Equation 2) (Fowler
2010).
The models themselves provide no information on the
influence of DNA organization on the generation of such
lesions. However, from simple considerations of biophysics,
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
3
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 5

larger chromosomes are more likely to participate in lesion
formation as they are larger targets for the creation of the ini-
tial DNA break (Cornforth et al. 2002). Nevertheless, both his-
toric and more recent data utilizing high throughput genome
analysis suggest that such simple assumptions may not fully
describe the earliest events in radiation-induced lethality. In
particular, as discussed further below, the organization of
DNA within the nucleus is not random, its organization into
separate functional domains, each consisting of looped DNA,
place a high degree of order on its packing within the
nuclear boundary (Lieberman-Aiden et al. 2009, Rao et al.
2014). Thus the effects of irradiation at the sub-nuclear level
may be illustrated using the tools and techniques of micro-
dosimetry placed in the context of the increasingly well-
defined
biological
organization
of
the
intact
eukaryote
nucleus.
Radiation induced lethal lesions are not generated
at random
To a first approximation, the generation of DNA breaks fol-
lowing ionizing irradiation would appear to be randomly dis-
tributed,
notwithstanding
the
greater
access
to
open
chromatin by free radicals (Chiu et al. 1986, Savage 1993,
Takata et al. 2013). After radiation exposure a variety of spe-
cies are produced from the radiolysis of water, the hydroxyl
radical (•OH) being the most frequent. Such radicals have
extremely short lifetimes, of the order of a nanosecond, corre-
sponding to a diffusion distance of 5 nm, and are capable
of damaging DNA if contact is made (Natarajan et al. 2010).
The generation of radiation-induced free radicals is respon-
sible for approximately two thirds of all DNA lesions and their
effects are more frequently observed within open regions of
the genome that are actively being transcribed, or are pre-
pared to do so (Chiu et al. 1986, Magnander et al. 2010,
Takata et al. 2013). This imposes a degree of selectivity on
the distribution of DNA damage following low Linear Energy
Transfer (LET) radiation exposure that will exhibit a bias to
those locations that are actively engaged in transcription, and
which are readily accessible to free radicals. Even if DNA
breaks themselves were assumed to be induced at random,
multiple lines of evidence suggest that their inappropriate
rejoining, such as translocations and other rearrangements,
are influenced by local genome architecture and post-break
processing (Engreitz et al. 2012, Hakim et al. 2012). Thus there
are inherent patterns to subsequent lesion formation. The
generation of rearrangements has historically been described
with reference to either the ‘contact first’ or ‘breakage first’
mechanisms (Aten et al. 2004, Meaburn et al. 2007). In the
contact first mechanism, the two chromosome elements are
in physical proximity prior to joint fragmentation – equivalent
in principle to ‘a type’ damage described in Equation (1),
where a cluster of ionization events may fragment more than
one DNA strand (Sutherland et al. 2000, Sankaranarayanan
et al. 2013). Alternatively, in the breakage first mechanism,
the breaks are generated as separate elements that may
require some mobility to meet and interact, broadly compat-
ible with ‘b type’ damage. Recent data has suggested that
the contact first mechanism may predominate in translocation
induction, though not to completely exclude break mobility
and subsequent interaction (Roukos et al. 2013). Here, to sim-
plify the discussion, the role of initial chromosome proximity
will be the focus in the genesis of aberrations.
Break proximity: Genome structure
Early studies on radiation-induced rearrangements identified
two common types of lethal rearrangements; centric rings,
formed by double-strand breaks on opposite arms of the
same chromosome, and dicentrics, from breaks on different
chromosomes (Figure 2A and B). The lesion may be expressed
subsequent to an attempt at cell division where such rear-
rangements restrict proper partitioning of genetic material
(Vakifahmetoglu et al. 2008). Though each type of rearrange-
ment requires two DNA double-strand breaks, careful micro-
scopic analysis of irradiated cells showed a smaller ratio of
dicentric to ring formation than would be predicted by
chance if all breaks interacted randomly (Sachs et al. 1997,
Sankaranarayanan et al. 2013). The explanation for such a
finding was assumed to be the physical localization of indi-
vidual chromosomes in their own ‘territories’, which favored
intra-chromosomal events (generating rings) over those that
required the interaction of two different chromosomes that
might be widely separated within the nucleus (Figure 2C).
This conclusion, reached from analysis of chromosome rear-
rangements in plant cells, was among the earliest observa-
tions suggesting the existence of chromosome territories in
the interphase nucleus and has been experimentally verified
many times since (Sax 1940, Cremer and Cremer 2001).
Break proximity: Low energy photoelectrons
The energy of photoelectrons generated within irradiated tis-
sue reach a maximum equal to that of the incident photon
minus its binding energy. Using this principle, very low
energy carbon K edge X-rays (0.28 keV), were employed as
discrete tools to initiate localized DNA damage due to the
very short path length (7 nm) of the photoelectrons gener-
ated. Using equipment that allowed the irradiation of single
cells, a three-fold enhancement in dicentric production was
observed, compared to 250 kVp X rays over a range of
1–4 Gy (Figure 2D) (Thacker et al. 1986, Sachs et al. 1997,
Griffin et al. 1998). In addition, the generation of aberrations
by such short-range electrons was found to be linearly related
to dose, D rather than D2, suggesting the two breaks required
were introduced as a single event (Sachs et al. 1997). This
implies an event (a type using the LQ formalism), occurring
in a small volume described by the range of the photoelec-
tron, was responsible for both breaks leading to the rear-
rangements
observed.
Thus
the
proximity
of
individual
chromosomes (or chromosome arms) were impacted by the
same induced photoelectron and/or associated free radical
production
or,
less
likely,
a
single
DNA
double-strand
break was able to initiate a rearrangement involving a prox-
imal,
undamaged,
chromosome
(Thacker
et
al.
1986,
Cornforth
1990,
Sachs
et
al.
1997,
Griffin
et
al.
1998,
4
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 6

Natarajan et al. 2010). In either case, however, both chromo-
some partners would likely need to be located close together
within the nucleus.
Break proximity: Ion pair and high LET irradiation
Similar proximity effects were found using cellular irradiation
with associated ion pairs (Geard 1985). In this approach,
molecular ion pairs are produced by stripping electrons from
accelerated diatomic species using Mylar films. When the
source diatomic material is deuterium (D2) the accelerated
particles are deuterons – one proton and one neutron.
Thicker Mylar films provide greater particle separation and
those particle pairs that were separated by 50 nm or less
were most effective at generating rearrangements (Geard
1985). Similarly to the data gathered using short-range photo-
electrons, this suggests that simultaneous fragmentation of
two DNA locations, generating ‘a’ type damage as defined
above (Equation 1), occurs efficiently in chromosome loca-
tions that are spatially related. Substantial data also exists on
the generation of chromosome rearrangements in relation to
LET. In particular, high LET radiations, such as alpha particles
or heavy ions, are up to 15-fold more efficient at generating
chromosome rearrangements than the same dose of low LET
radiation (Hada et al. 2011, Franken et al. 2012). These data
show that the number of DNA double-strand breaks induced
by such high LET irradiation, as measured by cH2AX (phos-
phorylated form of Histone H2A family member ‘X’) foci, are
similar to that seen after the same dose of low LET radiation,
suggesting it is the spatial distribution of the breaks, in add-
ition to their number, that influence aberration formation and
hence toxicity (Franken et al. 2012). Simplistically, a broadly
linear track across a cell nucleus from a high LET particle
places
the
majority
of
DNA
breaks
within
the
track
core itself – a cylinder of approximately 10 nm diameter
(Cucinotta et al. 1998). Here, due to spatial proximity and
increased concentration, these breaks are more likely to inter-
act than those scattered throughout the genome induced by
sparsely ionizing low LET radiation. Further, particles that
deliver 100 keV/lm energy, alpha particles and some neu-
tron energies, provide maximal cytotoxicity. This energy-
dependent cytotoxicity has been attributed to the inter-ion-
ization distance of such particles that matches the width of
the DNA alpha helix, optimizing the probability of a DNA
double-strand break (Hall and Giaccia 2012).
Direct measurement of lesions
The discussion above is derived from observations of lethal
rearrangements obtained primarily with discrete and non-
traditional radiation sources and shows that chromosome
proximity is a key element in rearrangement formation. To
specifically address the question of chromosome orientation
on the generation of rearrangements after conventional high
energy radiation exposures, a comprehensive study was initi-
ated, in a single human cell line, to screen for the frequency
of such rearrangements across all autosomes (Cornforth et al.
2002). In this analysis, chromosomes were uniquely labeled
using
multiplex
FISH
(mFISH
–
Fluorescence
In-Situ
Hybridization), such that rearrangement partners could be
assigned to specific chromosomes (Cornforth et al. 2002). As
expected, larger chromosomes, being larger radiation targets,
were damaged more frequently. Somewhat surprisingly, and
with few exceptions that included a group of centrally
located gene rich chromosomes (Chrs. 1, 16, 17, 19 and 22)
that showed increased mutual interactions, each chromosome
showed the potential to fuse with any other. This finding
appears to run counter to any role of chromosome proximity
in mediating rearrangement frequency. However, though a
comprehensive survey, such experiments carried out using
Figure 2. Dicentric and ring aberrations produced by irradiation lethal chromosome rearrangements produced by fusion two DNA double-strand breaks (dotted lines)
that are sub subsequently fused (arrows) either (A) between different chromosomes (dicentric), or (B) within the same chromosome (ring). (C) Chromosome breaks
(arrows) within the same chromosome have a higher probability of interacting, forming a ring structure, than breaks between chromosomes generating dicentrics or a
translocation. (D) Two chromosome elements that are physically adjacent are more likely to be fragmented by a single ionization cluster, and subsequently interact, a
process that is linearly related to dose, not dose squared.
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
5
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 7

chromosome specific FISH probes can only offer analytical
resolution at the level of whole chromosomes within a popu-
lation of cells. Thus individual chromosome territories, if they
vary in terms of their immediate contact partners from cell to
cell, will generate a random distribution of rearrangements
even if initial chromosome contact is the key element in
rearrangement formation. To dissect this aspect chromosome
organization on rearrangement probability, further technical
refinements were required. These advances are provided by
‘next generation’ or high throughput chromatin technologies,
which, as discussed below, offer base-pair-level sequencing
accuracy assayed at a resolution of a single cell.
Hierarchical organization of eukaryote DNA
The folding of 3  109 bp of linear DNA molecule into the cell
nucleus, approximating a sphere of diameter 8–15 lm, is clas-
sically described by the regular interaction of DNA with a his-
tone octamer, in the ‘beads on a string’ model. Through
increasing levels of compaction that involves a 30 nm fiber
containing 6–11 nucleosome elements per radial turn, chro-
matin eventually compacts into microscopically identifiable
chromosomes during mitosis (Hubner et al. 2013). Both the
‘beads on a string’ and chromosome models are likely accur-
ate descriptions of the two extremes of nuclear organization.
However, increasing doubt has been placed on the intermedi-
ate organization of DNA, mostly due to the as yet unresolved
degree to which in vitro artifacts perturb analysis (Fussner
et al. 2012, Nishino et al. 2012). Despite this important caveat,
a substantial body of work has addressed the question of
nuclear structure, using a range of different tools that pro-
vides some consensus. Using FISH labelling of sequential
tracts of DNA on the same chromosome it was shown that
the DNA backbone behaves as a flexible polymer, loosely
constrained within its own domain (Yokota et al. 1995). In
addition, by examination of the orientation of individual FISH
targets (loci closer than predicted by simple linear sequence)
it was deduced that the DNA was also organized in a series
of loops. Such a finding is supported by prior work that
inferred the presence of such constrained loops by their abil-
ity to relax and rewind their native supercoiling under the
influence of intercalating agents such as ethidium bromide, a
process completely inhibited by a break within the con-
strained loop (Cook and Brazell 1976, Khodarev et al. 1997).
Estimates of the loop size vary considerably, from < 0.1 Mbp
at the low end, with most however clustering in the 1–2.5
Mbp range, such differences likely linked to the difficulty in
working with such delicate structures and the specific tech-
nique used (Jackson et al. 1990, Khodarev et al. 1997,
Johnston et al. 1998, Ostashevsky et al. 1999). Nevertheless,
the not unexpected detection of a sub-chromosomal fine
structure within the nucleus has stimulated the generation of
biologic modeling strategies, this time focused on the impact
of such organization on lesion generation.
Rearrangements and genome proximity mapping
Next-generation sequencing (NGS) refers to the biochemical
and bioinformatic techniques that enable whole genomes to
be sequenced from experimentally fragmented DNA. A com-
bination of new NGS technology and experimental ingenuity
produced a method called chromosome conformation cap-
ture (3C), which has since been elaborated into many var-
iants, such as Hi-C (Engreitz et al. 2012, Dekker et al. 2013,
Koboldt et al. 2013). These techniques assay the spatial struc-
ture of the nuclear chromatin at a whole-genome scale, a
tool of particular relevance to radiation-induced aberrations.
In 3C, the three-dimensional architecture of the genome is
determined by chemically cross-linking genomic DNA, which
binds together segments of DNA that are physically adjacent
to each other at the moment of analysis. Cross-linked loci
may be on the same or different chromosomes. The genome
is then digested with a frequently cutting restriction endo-
nuclease, to fragment the genome and, in the case of Hi-C,
tagged with a biotin molecule at the fragmentation site
(Figure 3). In Hi-C, the biotin tag is then used to isolate the
individual chimeric molecules produced. Finally, using a dilute
concentration of DNA to enrich for intramolecular religation,
the circular ligation products formed are again fragmented,
and sequenced. With this system, the proximity of any spe-
cific genomic sequence to any other may be assessed by the
sequence composition of the chimeric fusions, generating a
Figure 3. Hi-C variant of Chromosome Conformation Capture. (A) Genome is cross-linked with formaldehyde, locations in close proximity are physically tethered to
each other. (B) Digesting the genome with a frequent (4 bp) restriction endonuclease generates free ends held together by the crosslink. (C) Each breaksite may be
tagged with biotin that is then ligated to its partner fragment (arrows), the entire construct cleaved into smaller fragments and the junctions themselves isolated by
pulling the biotin fragments out of solution. (D) DNA fragments containing biotin at the sequence junction now contain non-contiguous sequence elements that may
be identified when screened with next generation paired-end sequencing (arrows) technologies.
6
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 8

spatial proximity map for the entire genome (Lieberman-
Aiden et al. 2009, Rao et al. 2014).
Physical chromosome territories and chromosome
fine structure
Using Hi-C to interrogate the nuclear organization of human
lymphoblastoid cells, it was found that the most frequent
sites of contact for any location are sequences that are on
the same linear strand of DNA. The enhanced contact
observed decreases monotonically over 90 Mbp, interac-
tions within the same chromosome exceeding that observed
between different chromosomes, by 1–2 orders of magnitude
(Lieberman-Aiden et al. 2009). This therefore provides com-
prehensive support for the presence of chromosome territo-
ries – locations on the same chromosome that are spatially
close together as discussed above. Experiments such as these
provide the genomic documentation for why the ratio of
intra-chromosomal rearrangements (rings) to inter-chromo-
somal associations (dicentrics) varies from a simple random
distribution; the former are, on average, formed by elements
that are physically closer together prior to damage. Though
the study quoted only addresses a physical association of
DNA,
others
have
shown
genes
that
are
coordinately
expressed are often situated close to together on the same
chromosome (Caron et al. 2001, Cohen et al. 2000). The
detailed analysis of intra-chromosomal associations available
through Hi-C provides information on the organization of the
DNA strand that is most satisfactorily explained as a fractal
globule, at least over the range of approximately 0.5–90 Mbp
(Mirny 2011, Hahn and Kim 2013). The term ‘fractal’ used
here refers to a pattern that displays similarity independent
of scale, this type of fractal packing of linear DNA into 3-
dimensional space is known as a Hilbert curve (Lieberman-
Aiden et al. 2009). Such a structure, which does not contain
knots, would permit the rapid access to specific regions of
the genome, by simple unfolding, looping out and subse-
quent refolding, a level of packaging that would be appropri-
ate for rapid gene access (Figure 4). The presence of a fractal
globule polymer that persists over substantial (Mbp) distances
indicates that over short segments of DNA there must be
substantial flexibility and repetitive contact between DNA
sequences; with such contacts decreasing in a mathematically
predictable fashion as the distances between specific points
on the linear DNA strand increase. This behavior may under-
pin the documented potency of radiation to generate large
scale chromosomal deletions, a process likely to be facilitated
if two adjacent points on the same chromosome are in phys-
ical contact when both are impacted by an ionization event
(Sankaranarayanan et al. 2013).
A practical example of this level of organization may
already have been observed in papillary thyroid cancer
where fusion between RET (REarranged during Transfection)
and either of its two frequent fusion partners, NCOA4
(Nuclear receptor COActivator 4) and CCDC6 (Coiled-Coil
Domain Containing 6), is strongly linked to the disease.
These three genes are all located within 18 Mbp on the
long
arm
of
chromosome
10
(Nikiforov
et
al.
1999,
Nikiforova et al. 2000). Fusion of RET to either of these
partners was commonly observed in those children devel-
oping thyroid tumors after the release of radioactivity, not-
ably I-131, from the Chernobyl reactor explosion (Williams
2008). Careful matching of individual breakpoints in RET
and NCOA4, separated by 8 Mbp, from patient tumors sug-
gested that both genes were located in a fixed position
prior to their joint fragmentation and subsequent fusion
(Figure 5) (Nikiforov et al. 1999). It is possible (though
unknown) that a fractal globule level of organization would
reach
a
common
solution
in
terms
of
unpacking
and
repacking a single region of DNA in a reproducible orienta-
tion, such that RET and NCOA4 are repetitively placed in
juxtaposition, at least in the thyroid. This may specifically
be the case however if both genes were transcribing at
the same time, acting in response to similar signals or
looped together via regulatory interactions (Cohen et al.
2000, Fullwood et al. 2009, Rao et al. 2014). This would
make them statistically favored candidates for joint frag-
mentation
and
rearrangement
subsequent
to
radiation
induced fragmentation. The link to transcription was made
more intriguing by an examination of RET and CCDC6 loca-
tions with FISH. Here, in an in vitro setting, RET and CCDC6
were found to be in close proximity in thyroid cells, but
less so in cells of other types, suggesting a cell specific
x
5´                                       3´
(A)
(B)
(C)
Figure 4. Chromatin organization interpreted from Hi-C experiments. (A) Chromosomes are localized in territories (three shown) however with substantial interdigita-
tion or physical overlap into other chromosome territories. (B) The locations of inter-chromosome (and intra-chromosome) contact is linked through sharing transcrip-
tional status; either on (arrow) or off (cross). (C) At the level of a discrete linear DNA strand the DNA is organized and efficiently packed within the genome as a
fractal globule where short-range self-associations predominate. DNA self-association is thus greatest over short distances (bp to Kbp) and reduces monotonically as
the distance (90 Mbp) increases. Diagrams adapted from concepts developed by Lieberman-Aiden et al. (2009).
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
7
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 9

and potentially transcription specific localization (Nikiforova
et al. 2000).
Functional chromosome territories
Using the same data set generated by Hi-C as discussed
above, Lieberman-Aiden also showed that the genome can
be compartmentalized into two broadly identifiable regions
that are self-associated, likely equivalent to euchromatin and
heterochromatin (Lieberman-Aiden et al. 2009). The link to
actively transcribing DNA was made by cross referencing Hi-C
proximity data to both transcriptional activity, DNAse I sensi-
tivity and levels of DNA methylation within Histone H3 lysine
36 (H3K36) that are associated with gene activity (Mikkelsen
et al. 2007). Such associations transcend the linear organiza-
tion of DNA, placing regions of gene activity into one physic-
ally interacting compartment, and heterochromatic DNA in
another. This may seem a trivial reiteration of the well-known
composition of chromatin. However, these data extend the
description of these regions as viewed by microscopy to
show, on a genome wide scale, that actively transcribing
regions from the same and different chromosomes are in
close physical contact; these are the same regions that are
also more prone to attack by radiation-induced free radicals
(Chiu et al. 1986, Cowell et al. 2007, Magnander et al. 2010,
Vasireddy et al. 2010, Takata et al. 2013). The compartmental-
ized proximity of such damage may facilitate the subsequent
rearrangement within compartments rather than between
them (Cornforth et al. 2002, Lieberman-Aiden et al. 2009).
Functional compartmentalization as described was ana-
lyzed in more detail with a particular emphasis on intra-
chromosomal contacts (Kalhor et al. 2012). Here, Kalhor and
colleagues, using a variant of the Hi-C procedure executed on
a solid matrix rather than in solution, found that for the larg-
est chromosomes (Chrs. 1–6, 8 and 10), intra-chromosomal
interactions across the centromere were restricted for those
genomic regions that were composed of transcriptionally
inactive chromatin. Actively transcribing regions were not so
restricted and exhibited a higher contact frequency over a
greater range of linear DNA sequence than non-transcribing
elements. These data therefore suggest that, for these chro-
mosomes, contact across the centromere, and any subse-
quent rearrangement such as a ring structure if fragmented,
would likely be restricted to regions containing actively tran-
scribing genes. These data parallel a similar whole genome
survey examining translocations where experimental breaks
inserted by the meganuclease, I-SceI into unique genes were
commonly fused to partner locations that were actively tran-
scribing (Chiarle et al. 2011).
Population-based analysis of the genome
Though a very useful approach, the 3C technique as dis-
cussed can only provide an average spatial mapping of speci-
fied locations across an entire population of cells. It is unclear
whether such organization is uniformly maintained within
defined cell lines, or potentially oligoclonal human tumor sys-
tems, or even between neighboring cells of the same geno-
type. To address this question, Kalhor subjected multiple 3C
datasets to population based analysis. Here the genome was
allocated into 428 chromatin blocks that contained similar
contact profiles with the rest of the genome (Kalhor et al.
2012). These were then tested through 10,000 genome simu-
lations to optimize fitting of the ‘block’ data with its inter-
action with the rest of the genome. This presentation of the
data suggested that multiple genomic conformations pro-
vided the best fit for the 3C data obtained within an other-
wise identical cell line. To address the question of variability
directly, Nagano and colleagues subjected a series of other-
wise identical single cells to Hi-C analysis (Nagano et al.
2013). Using mouse splenic CD4 (Cluster of Differentiation 4
or T helper) cells, they found substantial variability between
inter-chromosomal contacts between the different individual
cells that were examined. Individual cells were found to have
relatively few chromosome contacts. These data suggest that
previous maps showing a high frequency of genomic inter-
action, and multiple points of contact, represent an averaging
of discrete data generated from uniquely organized single
cells. This implies that even within a single cell type, the gen-
ome adopts multiple unique orientations, perhaps linked to
cell cycle stage, tissue type or transcriptional profile such as
that controlling stem cell potential (Charafe-Jauffret et al.
2009, 2013, de Wit et al. 2013, Aranda-Anzaldo et al. 2014).
Extrapolating these findings to the mechanism whereby indi-
vidual chromosomes fuse suggests that any specific chromo-
some contacts between unique chromosomes may be lost in
the noise of averaging multiple cell interactions.
These data are significant in terms of the generation of
radiation-induced rearrangements in that it questions their
broadly random generation, as shown by mFISH studies
(Cornforth et al. 2002). Here, the argument for randomness
was made by examining an entire population of irradiated
cells, showing that each chromosome has a broadly equal
opportunity to interact with any other. The data discussed
above brings this interpretation into question, suggesting
NCOA4
RET
0  
500
1680 bp
1500
1000
0  
500
1840 bp
1500
1000
40
45
50
55
60
65 Mbp
< cent
Chr10
RET
NCOA4
CCDC6
Figure 5. Proximity mapping of breakpoints in papillary thyroid cancer, redrawn
using data from Nikiforov et al. (1999). (Top) Shown are the breakpoints mapped
in specific segments of intron 11 and intron 5 of the RET-NCOA4 (REarranged
during Transfection – Nuclear receptor COActivator 4) gene fusion respectively,
and shown to scale. Here 9/12 breakpoints share a broadly parallel relationship
with each other (bold lines), as do the 3/12 remaining breakpoints (faint lines).
Such arrangements would be generated if both genes are located in a reprodu-
cible relationship with each other, such that a single ionization event affects
both genes simultaneously. (Bottom) Location and transcriptional direction of
each gene on chromosome 10. Despite the linear separation the data are consist-
ent with reproducible gene contacts during radiation-associated cleavage that
arise from a single radiation event.
8
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 10

that within a single cell type, multiple genomic states may be
generated, each of which brings together specific genomic
components to execute a required function (Fullwood et al.
2009, Sandhu et al. 2012). This will provide differing probabil-
ities, among single cells, for simultaneous DNA fragmentation
and subsequent interaction of the regions that are uniquely
in contact. For cell killing by irradiation it is the impact on sin-
gle cells that is the key outcome, reflected in the adoption of
the clonogenic (single cell) assay as the ‘gold standard’ for
toxicity testing. There may be multiple factors that drive the
specific organization of genomic components; and the role of
transcription has already been implicated in such structural
reorganization as a source of more ‘open’ and free radical
accessible chromatin. Clearly for tumor eradication it is the
response of individual cells, particularly those with an infinite
reproductive future, which is of most concern.
Transcription linked genome organization
The advent of NGS and 3C technology has uncovered a
complex interacting web of both short and long range (inter-
chromosomal) interactions that sustain the cells transcrip-
tional profile and execute tissue specific functions (Sandhu
et al. 2012). These networks involve the physical association
of widely separated (in terms of assigned chromosome) gen-
ome locations that may preferentially interact if jointly frag-
mented
by
irradiation.
Such
an
association
has
been
demonstrated using the I-SceI meganuclease as a surrogate
for radiation, a widely used system for fragmenting defined
genomic locations (Richardson and Jasin 2000, Chiarle et al.
2011, Schipler and Iliakis 2013). With this system targeting
either c-myc (cellular myelocytomatosis oncogene) or IgH
(Immunoglobulin H) genes, it was found that the experimen-
tally fragmented genes preferentially underwent rearrange-
ment with actively transcribing genes, particularly the histone
depleted transcriptional start sites (Chiarle et al. 2011). These
specific rearrangements are likely to occur, at least in part,
due to the proximity of the fragmented gene and its subse-
quent partner, perhaps localized by sharing a common func-
tion, such as transcription. As an example, the TMPRSS2/ERG
(TransMembrane PRoteaSe, Serine 2/Erythroblast transform-
ation-specific Related Gene) gene fusion, observed in prostate
tumors of 50% of men with this disease, may be generated
by androgen mediated co-localization of the two genes, in
this case located on the same chromosome (Clark and
Cooper 2009, Lin et al. 2009). The subsequent addition of a
small radiation dose facilitates their fragmentation and fusion
(Lin et al. 2009). The androgen mediated co-localization of
these genes provides support for sub-nuclear functional com-
partments described as transcription factories, where genes
with linked function aggregate in response to discrete tran-
scriptional stimuli (Osborne et al. 2007, Fullwood et al. 2009).
The presence of such factories was also inferred from the
co-localization of the murine c-myc and IgH genes, on chro-
mosomes 15 and 12 respectively, which are common trans-
location partners in plasmacytomas and Burkitts lymphoma
(Osborne et al. 2007). Thus it is apparent that transcriptional
programs are involved in defining the higher order structure
of the genome and such functions may influence the contact
proximity of both transcribing genes and their promoter/
enhancer
interactions,
if
appropriate
stimuli
are
applied
(Fullwood et al. 2009). In support of the general role of gen-
ome proximity in generating translocations, chromosome/
chromosome proximity has been directly linked to the poten-
tial to undergo translocations following irradiation, at least in
a murine system (Zhang et al. 2012). In addition, the Hi-C
contact database generated by Lieberman-Aiden was cross-
referenced with a large set of clinically relevant translocations,
including the Mittelman database of rearrangements (Engreitz
et al. 2012). Here, a strong correlation was found showing
that for multiple clinically relevant translocations the partner
genes are preferentially closely associated, prior to any fusion.
It is a logical extension that genomic regions in contact may
be targets for lethal fusion formation if both are damaged by
irradiation, or other genotoxic agents. Though only transloca-
tions are discussed in this section, as they may be observed
in surviving clones of cells, rearrangements that have the
potential to be lethal, such as a dicentric, differ only in the
inclusion of a second centromere (Meaburn et al. 2007).
The process that drives the physical association of potential
partners and the subsequent formation of the lesion may be
very similar, thus supporting the ‘contact first’ hypothesis for
radiation induced rearrangements.
Transcriptional
upregulation
may
therefore
have
two,
potentially synergistic, effects on radiation-induced lesion for-
mation. First, regions that are transcriptionally active are more
likely to undergo radiation-induced free radical damage due
to their relaxed conformation facilitating free radical access.
Second, transcriptionally active genes and their promoter/
enhancer targets may preferentially stay in contact, making
them susceptible to simultaneous fragmentation from either
free radicals or direct ionization. As an example, the ERG tran-
scription factor that is commonly over expressed in a range
of tumors, was subject of an in-depth analysis of its effect on
genomic
organization,
using
3C
techniques.
Here,
over
expression of ERG showed reproducible alterations in the spa-
tial organization of the genome, a process that was consistent
with the activation of genes linked to its role in oncogenic
development (Rickman et al. 2012). Perhaps of greater inter-
est however is the study of pluripotent stem cells, a cell type
implicated as the key driver cell in the maintenance of a
viable tumor mass and a logical target for therapeutic attack
(Al-Hajj et al. 2003, Singh et al. 2004). It was found that the
binding sites of those factors linked to the stem cell state,
such as Nanog and Oct4, were preferentially associated, gen-
erating a stem cell specific set of chromosome contacts (de
Wit et al. 2013, Ay et al. 2014). Both examples were generated
from population based analyses, indicating the commonality
of the associations found.
Relevance to radiation biology and treatment
The discussion above identifies transcriptionally active regions
of the genome as more likely to be fragmented by ionizing
radiation, specifically through free radical access, and also
more
likely
to
fuse
when
broken
due
to
their
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
9
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 11

compartmentalized proximity. Placed in the language of radi-
ation biology and the LQ formalism, such events may corres-
pond to ‘a’ type damage where fragmentation occurs in two
chromosomes simultaneously and is linearly proportional to
dose, rather than separately identifiable events proportional
to dose squared, as events induced by separate ionizations
interact (Fowler 2010). This interpretation may however be
too simplistic as experimental data shows that most survival
curves tend to adhere to a simple exponential at higher
doses, not the continuous curve as implied by Equation (2)
(Park et al. 2008). It follows therefore that at high doses, per-
haps >10 Gy, lethal lesions are generated by single events,
such as the simultaneous fragmentation of two chromosomes,
rather than the interaction of separate sub-lesions that pro-
vides the curvature to the LQ response. If this is correct then
spatial contacts within the genome will have an influence on
what specific rearrangements will be most likely to occur dur-
ing the transition from low to very high single doses.
Specifically, as the dose and number of breaks increase, the
probability of two breaks interacting within the same nuclear
compartment
as
defined
by
transcriptional
status,
also
increases. This is particularly relevant as cell-type specific tran-
scriptional programs activate their complement of genes, a
feature that is a signature corruption within the transformed
phenotype (Lee and Young 2013).
In terms of practical utility, a key question is whether
knowledge of DNA organization at the level discussed here
provides any useful information to the radiobiologist or radi-
ation oncologist. Transcriptional programs that enhance spe-
cific chromosome contact likely suppress others, the potential
net effect being a ‘zero sum’ of site-restricted lethal aberra-
tions with no change in absolute lethality. Alternatively, cell-
specific transcriptional programs, such as those characteristic
of individual tissues and/or tumors, may generate a spatial
proximity interactome that offers a common physical target
for the insertion of DNA breaks and their subsequent process-
ing. Though the focus in this review is the spatial organiza-
tion of chromatin, the formation of any such lethal lesions
will depend on the recruitment and effective action of DNA
damage repair programs, such as NHEJ and HR, recently the
subject of a comprehensive review in the context of radiation
damage (Thompson 2012). By definition, if such programs
operate efficiently no lethal lesions will occur as the chromo-
some
integrity
and
orientation
is
unchanged.
It
is
the
inappropriate ligation of two disparate chromosome frag-
ments that are in close physical contact that generates
lesions. It follows that the contribution of either corrupt or
intact DNA repair programs present in human cancers to radi-
ation lethality may be impacted by the spatial organization of
the DNA breaks they act upon, not just their ability to rejoin
naked DNA templates.
To demonstrate the clinical utility of a more nuanced view
of nuclear organization, it is important to identify any mech-
anistic relationship between the spatial arrangement of the
genome pre-irradiation and the rearrangements produced by
DNA repair programs post-irradiation. Does the chromatin
state of the cell pre-irradiation bias towards certain structural
rejoining events? If such a role is observed, then an unex-
plored avenue of radiation treatment sensitization may be
the judicious administration of transcriptome-modulating sys-
tems (such as, RNA interference or microRNA) to manipulate
specific genetic programs within the cancerous tissue before
treatment.
The following questions need to be addressed:
1. Does chromosome contact in individual cells vary between
cell
types
(different
tissues,
as
well
as
normal
vs.
transformed)?
2. What controls the association of chromosome contacts in
individual cells? How do these factors interact and influ-
ence each other?
3. If, as suggested from current Hi-C data in single cells,
chromosome contact is restricted; do radiation-induced
aberrations occur predominately within this class of physic-
ally associated chromosome contacts?
4. How does the spatial location of individual DNA breaks
impact the efficiency of their restitution by the known
repair pathways?
5. What types of therapeutic approaches could alter the
nuclear architecture of cancerous tissue in a way that
would increase the efficacy of radiation treatment?
Resolving the above questions will likely require the use of
single cell Hi-C technology merged with aberration analysis in
irradiated material, a challenging application, which is now
technically feasible. The effects on single cells is the preferred
platform for experiment as the variability of lesions within
irradiated populations are difficult to tease out using current
NGS systems. Such questions will determine whether the gen-
eration of lethal aberrations may be manipulated, for eventual
therapeutic gain, through targeted alterations of chromosome
contacts in normal tissues, tumors, or both.
Disclosure statement
The authors report no conflict of interest. The authors alone
are responsible for the content and writing of the paper.
References
Adachi N, Ishino T, Ishii Y, Takeda S, Koyama H. 2001. DNA ligase IV-defi-
cient cells are more resistant to ionizing radiation in the absence of
Ku70: Implications for DNA double-strand break repair. Proc Natl Acad
Sci USA. 98:12109–12113.
Al-Hajj M, Wicha MS, Benito-Hernandez A, Morrison SJ, Clarke MF. 2003.
Prospective identification of tumorigenic breast cancer cells. Proc Natl
Acad Sci USA. 100:3983–3988.
Aranda-Anzaldo A, Dent MA, Martinez-Gomez A. 2014. The higher-order
structure in the cells nucleus as the structural basis of the post-mitotic
state. Prog Biophys Mol Biol. 114:137-145.
Aten JA, Stap J, Krawczyk PM, Van Oven CH, Hoebe RA, Essers J, Kanaar
R. 2004. Dynamics of DNA double-strand breaks revealed by clustering
of damaged chromosome domains. Science. 303:92–95.
Averbeck NB, Ringel O, Herrlitz M, Jakob B, Durante M, Taucher-Scholz G.
2014. DNA end resection is needed for the repair of complex lesions
in G1-phase human cells. Cell Cycle. 13:2509–2516.
Ay F, Bailey TL, Noble WS. 2014. Statistical confidence estimation for Hi-C
data reveals regulatory chromatin contacts. Genome Res. 24:999-101.
Ballarini F. 2010. From DNA radiation damage to cell death: Theoretical
approaches. J Nucleic Acids. 2010:350608.
10
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 12

Brown JM, Attardi LD. 2005. The role of apoptosis in cancer development
and treatment response. Nat Rev Cancer. 5:231–237.
Caron H, Van Schaik B, Van Der Mee M, Baas F, Riggins G, Van Sluis P,
Hermus MC, Van Asperen R, Boon K, Voute PA, et al. 2001. The human
transcriptome map: Clustering of highly expressed genes in chromo-
somal domains. Science. 291:1289–1292.
Chadwick KH, Leenhouts HP. 1973. A molecular theory of cell survival.
Phys Med Biol. 18:78–87.
Chapman JR, Taylor MR, Boulton SJ. 2012. Playing the end game: DNA
double-strand break repair pathway choice. Mol Cell. 47:497–510.
Charafe-Jauffret E, Ginestier C, Bertucci F, Cabaud O, Wicinski J, Finetti P,
Josselin E, Adelaide J, Nguyen TT, Monville F, et al. 2013. ALDH1-posi-
tive cancer stem cells predict engraftment of primary breast tumors
and are governed by a common stem cell program. Cancer Res.
73:7290–7300.
Charafe-Jauffret E, Ginestier C, Iovino F, Wicinski J, Cervera N, Finetti P,
Hur MH, Diebel ME, Monville F, Dutcher J, et al. 2009. Breast cancer
cell lines contain functional cancer stem cells with metastatic capacity
and a distinct molecular signature. Cancer Res. 69:1302–1313.
Chiarle R, Zhang Y, Frock RL, Lewis SM, Molinie B, Ho YJ, Myers DR, Choi
VW, Compagno M, Malkin DJ, et al. 2011. Genome-wide translocation
sequencing reveals mechanisms of chromosome breaks and rearrange-
ments in B cells. Cell. 147:107–119.
Chiu SM, Friedman LR, Xue LY, Oleinick NL. 1986. Modification of DNA
damage in transcriptionally active vs. bulk chromatin. Int J Radiat
Oncol Biol Phys. 12:1529–1532.
Clark JP, Cooper CS. 2009. ETS gene fusions in prostate cancer. Nat Rev
Urol. 6:429–439.
Cohen BA, Mitra RD, Hughes JD, Church GM. 2000. A computational ana-
lysis of whole-genome expression data reveals chromosomal domains
of gene expression. Nat Genet. 26:183–186.
Cook PR, Brazell IA. 1976. Conformational constraints in nuclear DNA.
J Cell Sci. 22:287–302.
Cornforth MN. 1990. Testing the notion of the one-hit exchange. Radiat
Res. 121:21–27.
Cornforth MN, Greulich-Bode KM, Loucas BD, Arsuaga J, Vazquez M, Sachs
RK,
Bruckner
M,
Molls
M,
Hahnfeldt
P,
Hlatky
L,
et
al.
2002.
Chromosomes are predominantly located randomly with respect to
each other in interphase human cells. J Cell Biol. 159:237–244.
Costes SV, Ponomarev A, Chen JL, Nguyen D, Cucinotta FA, Barcellos-Hoff
MH. 2007. Image-based modeling reveals dynamic redistribution of
DNA damage into nuclear sub-domains. PLoS Comput Biol. 3:e155.
Cowell IG, Sunter NJ, Singh PB, Austin CA, Durkacz BW, Tilby MJ. 2007.
gammaH2AX foci form preferentially in euchromatin after ionising-radi-
ation. PLoS One. 2:e1057.
Cremer T, Cremer C. 2001. Chromosome territories, nuclear architecture
and gene regulation in mammalian cells. Nat Rev Genet. 2:292–301.
Cucinotta FA, Nikjoo H, Goodhead DT. 1998. The effects of delta rays on
the number of particle-track traversals per cell in laboratory and space
exposures. Radiat Res. 150:115–119.
De Wit E, Bouwman BA, Zhu Y, Klous P, Splinter E, Verstegen MJ, Krijger
PH, Festuccia N, Nora EP, Welling M, et al. 2013. The pluripotent gen-
ome in three dimensions is shaped around pluripotency factors.
Nature. 501:227–231.
Dekker J, Marti-Renom MA, Mirny LA. 2013. Exploring the three-dimen-
sional organization of genomes: Interpreting chromatin interaction
data. Nat Rev Genet. 14:390–403.
Edwards AA, Moiseenko VV, Nikjoo H. 1996. On the mechanism of the for-
mation of chromosomal aberrations by ionising radiation. Radiat
Environ Biophys. 35:25–30.
Engreitz JM, Agarwala V, Mirny LA. 2012. Three-dimensional genome
architecture influences partner selection for chromosomal transloca-
tions in human disease. PLoS One. 7:e44196.
Fowler JF. 2010. 21 years of biologically effective dose. Br J Radiol.
83:554–568.
Franken NA, Hovingh S, Ten Cate R, Krawczyk P, Stap J, Hoebe R, Aten J,
Barendsen GW. 2012. Relative biological effectiveness of high linear
energy transfer alpha-particles for the induction of DNA-double-strand
breaks, chromosome aberrations and reproductive cell death in SW-
1573 lung tumour cells. Oncol Rep. 27:769–774.
Friedland W, Dingfelder M, Kundrat P, Jacob P. 2011. Track structures,
DNA targets and radiation effects in the biophysical Monte Carlo simu-
lation code PARTRAC. Mutat Res. 711:28–40.
Friedland W, Kundrat P. 2013. Track structure based modelling of chromo-
some aberrations after photon and alpha-particle irradiation. Mutat
Res. 756:213–223.
Fullwood MJ, Liu MH, Pan YF, Liu J, Xu H, Mohamed YB, Orlov YL,
Velkov S, Ho A, Mei PH, et al. 2009. An oestrogen-receptor-alpha-
bound human chromatin interactome. Nature. 462:58–64.
Fussner E, Strauss M, Djuric U, Li R, Ahmed K, Hart M, Ellis J, Bazett-Jones
DP. 2012. Open and closed domains in the mouse genome are config-
ured as 10-nm chromatin fibres. EMBO Rep. 13:992–996.
Geard CR. 1985. Charged particle cytogenetics: Effects of LET, fluence,
and particle separation on chromosome aberrations. Radiat Res (Suppl.
8):S112–121.
Goodhead DT. 2006. Energy deposition stochastics and track structure:
What about the target? Radiat Prot Dosimetry. 122:3–15.
Goodhead DT, Thacker J, Cox R. 1979. Effectiveness of 0.3 keV carbon
ultrasoft X-rays for the inactivation and mutation of cultured mamma-
lian cells. Int J Radiat Biol Relat Stud Phys Chem Med. 36:101–114.
Griffin CS, Hill MA, Papworth DG, Townsend KM, Savage JR, Goodhead
DT. 1998. Effectiveness of 0.28 keV carbon K ultrasoft X-rays at produc-
ing simple and complex chromosome exchanges in human fibroblasts
in vitro detected using FISH. Int J Radiat Biol. 73:591–598.
Hada M, Wu H, Cucinotta FA. 2011. mBAND analysis for high- and low-
LET radiation-induced chromosome aberrations: A review. Mutat Res.
711:187–192.
Hahn S, Kim D. 2013. Physical origin of the contact frequency in chromo-
some conformation capture data. Biophys J. 105:1786–1795.
Hakim O, Resch W, Yamane A, Klein I, Kieffer-Kwon KR, Jankovic M,
Oliveira T, Bothmer A, Voss TC, Ansarah-Sobrinho C, et al. 2012. DNA
damage defines sites of recurrent chromosomal translocations in B
lymphocytes. Nature. 484:69–74.
Hall EJ, Giaccia AJ. 2012. Radiobiology for the radiologist. Philadelphia:
Wolters Kluwer Health.
Herr L, Friedrich T, Durante M, Scholz M. 2014. A model of photon cell
killing based on the spatio-temporal clustering of DNA damage in
higher order chromatin structures. PLoS One. 9:e83923.
Hubner MR, Eckersley-Maslin MA, Spector DL. 2013. Chromatin organiza-
tion and transcriptional regulation. Curr Opin Genet Dev. 23:89–95.
Iliakis G. 1991. The role of DNA double strand breaks in ionizing radi-
ation-induced killing of eukaryotic cells. Bioessays. 13:641–648.
Jackson DA, Dickinson P, Cook PR. 1990. The size of chromatin loops in
HeLa cells. EMBO J. 9:567–571.
Jackson SP, Bartek J. 2009. The DNA-damage response in human biology
and disease. Nature. 461:1071–1078.
Jaffray DA. 2012. Image-guided radiotherapy: From current concept to
future perspectives. Nat Rev Clin Oncol. 9:688–699.
Johnston PJ, MacPhail SH, Banath JP, Olive PL. 1998. Higher-order chro-
matin
structure-dependent
repair
of
DNA
double-strand
breaks:
Factors
affecting
elution
of
DNA
from
nucleoids.
Radiat
Res.
149:533–542.
Kalhor R, Tjong H, Jayathilaka N, Alber F, Chen L. 2012. Genome architec-
tures revealed by tethered chromosome conformation capture and
population-based modeling. Nat Biotechnol. 30:90–98.
Kavanagh JN, Redmond KM, Schettino G, Prise KM. 2013. DNA double
strand break repair: A radiation perspective. Antioxid Redox Signal.
18:2458–2472.
Kellerer AM, Rossi HH. 2012. A generalized formulation of dual radiation
action. Radiat Res. 178:Av204–213.
Khodarev
NN,
Narayana
A,
Constantinou
A,
Vaughan
AT.
1997.
Topologically constrained domains of supercoiled DNA in eukaryotic
cells. DNA Cell Biol. 16:1051–1058.
Koboldt DC, Steinberg KM, Larson DE, Wilson RK, Mardis ER. 2013. The
next-generation sequencing revolution and its impact on genomics.
Cell. 155:27–38.
Lea DE. 1955. Actions of radiation on living cells. New York: Cambridge
University Press.
Lee TI, Young RA. 2013. Transcriptional regulation and its misregulation in
disease. Cell. 152:1237–1251.
INTERNATIONAL JOURNAL OF RADIATION BIOLOGY
11
Downloaded by [University of California, San Diego] at 06:07 14 March 2016

## Page 13

Lieberman-Aiden E, Van Berkum NL, Williams L, Imakaev M, Ragoczy T,
Telling A, Amit I, Lajoie BR, Sabo PJ, Dorschner MO, et al. 2009.
Comprehensive mapping of long-range interactions reveals folding
principles of the human genome. Science. 326:289–293.
Lin C, Yang L, Tanasa B, Hutt K, Ju BG, Ohgi K, Zhang J, Rose DW, Fu XD,
Glass CK, et al. 2009. Nuclear receptor-induced chromosomal proximity
and DNA breaks underlie specific translocations in cancer. Cell.
139:1069–1083.
Magnander K, Hultborn R, Claesson K, Elmroth K. 2010. Clustered DNA
damage in irradiated human diploid fibroblasts: Influence of chromatin
organization. Radiat Res. 173:272–282.
McVey M, Lee SE. 2008. MMEJ repair of double-strand breaks (director’s
cut):
Deleted
sequences
and
alternative
endings.
Trends
Genet.
24:529–538.
Meaburn KJ, Misteli T, Soutoglou E. 2007. Spatial genome organization in
the formation of chromosomal translocations. Semin Cancer Biol.
17:80–90.
Mikkelsen TS, Ku M, Jaffe DB, Issac B, Lieberman E, Giannoukos G, Alvarez
P, Brockman W, Kim TK, Koche RP,et al. 2007. Genome-wide maps of
chromatin state in pluripotent and lineage-committed cells. Nature.
448:553–560.
Mirny LA. 2011. The fractal globule as a model of chromatin architecture
in the cell. Chromosome Res. 19:37–51.
Nagano T, Lubling Y, Stevens TJ, Schoenfelder S, Yaffe E, Dean W, Laue
ED, Tanay A, Fraser P. 2013. Single-cell Hi-C reveals cell-to-cell variabil-
ity in chromosome structure. Nature. 502:59–64.
Natarajan AT, Palitti F, Hill MA, Stevens DL, Ahnstrom G. 2010. Influence
of DMSO on Carbon K ultrasoft X-rays induced chromosome aberra-
tions in V79 Chinese hamster cells. Mutat Res. 691:23–26.
Nikiforov YE, Koshoffer A, Nikiforova M, Stringer J, Fagin JA. 1999.
Chromosomal breakpoint positions suggest a direct role for radiation
in inducing illegitimate recombination between the ELE1 and RET
genes
in
radiation-induced
thyroid
carcinomas.
Oncogene.
18:6330–6334.
Nikiforova MN, Stringer JR, Blough R, Medvedovic M, Fagin JA, Nikiforov
YE. 2000. Proximity of chromosomal loci that participate in radiation-
induced rearrangements in human cells. Science. 290:138–141.
Nishino Y, Eltsov M, Joti Y, Ito K, Takata H, Takahashi Y, Hihara S,
Frangakis AS, Imamoto N, Ishikawa T, et al. 2012. Human mitotic chro-
mosomes consist predominantly of irregularly folded nucleosome
fibres without a 30-nm chromatin structure. EMBO J. 31:1644–1653.
Nomiya T. 2013. Discussions on target theory: Past and present. J Radiat
Res. 54:1161–1163.
O’Neill P, Wardman P. 2009. Radiation chemistry comes before radiation
biology. Int J Radiat Biol. 85:9–25.
Obe G, Johannes C, Ritter S. 2010. The number and not the molecular
structure of DNA double-strand breaks is more important for the for-
mation
of
chromosomal
aberrations:
A
hypothesis.
Mutat
Res.
701:3–11.
Olive PL. 1998. The role of DNA single- and double-strand breaks in cell
killing by ionizing radiation. Radiat Res. 150:S42–51.
Osborne CS, Chakalova L, Mitchell JA, Horton A, Wood AL, Bolland DJ,
Corcoran AE, Fraser P. 2007. Myc dynamically and preferentially relo-
cates to a transcription factory occupied by Igh. PLoS Biol. 5:e192.
Ostashevsky JY, Reichman B, Lange CS. 1999. Higher-order structure of
mammalian chromatin deduced from viscoelastometry data. J Biomol
Struct Dyn. 17:567–580.
Park C, Papiez L, Zhang S, Story M, Timmerman RD. 2008. Universal sur-
vival curve and single fraction equivalent dose: Useful tools in under-
standing potency of ablative radiotherapy. Int J Radiat Oncol Biol Phys.
70:847–852.
Pfeiffer P, Goedecke W, Kuhfittig-Kulle S, Obe G. 2004. Pathways of DNA
double-strand break repair and their impact on the prevention and
formation
of
chromosomal
aberrations.
Cytogenet
Genome
Res.
104:7–13.
Price BD, D’Andrea AD. 2013. Chromatin remodeling at DNA double-
strand breaks. Cell. 152:1344–1354.
Rao SS, Huntley MH, Durand NC, Stamenova EK, Bochkov ID, Robinson JT,
Sanborn AL, Machol I, Omer AD, Lander ES, et al. 2014. A 3D map of
the human genome at kilobase resolution reveals principles of chro-
matin looping. Cell. 159:1665–1680.
Richardson C, Jasin M.
2000. Frequent chromosomal translocations
induced by DNA double-strand breaks. Nature. 405:697–700.
Rickman DS, Soong TD, Moss B, Mosquera JM, Dlabal J, Terry S,
MacDonald TY, Tripodi J, Bunting K, Najfeld V, et al. 2012. Oncogene-
mediated alterations in chromatin conformation. Proc Natl Acad Sci
USA. 109:9083–9088.
Roukos V, Voss TC, Schmidt CK, Lee S, Wangsa D, Misteli T. 2013. Spatial
dynamics
of
chromosome
translocations
in
living
cells.
Science.
341:660–664.
Sachs RK, Chen AM, Brenner DJ. 1997. Review: Proximity effects in the
production of chromosome aberrations by ionizing radiation. Int J
Radiat Biol. 71:1–19.
Sandhu KS, Li G, Poh HM, Quek YL, Sia YY, Peh SQ, Mulawadi FH,
Lim J, Sikic M, Menghi F, et al. 2012. Large-scale functional organ-
ization
of
long-range
chromatin
interaction
networks.
Cell
Rep.
2:1207–1219.
Sankaranarayanan K, Taleei R, Rahmanian S, Nikjoo H. 2013. Ionizing radi-
ation and genetic risks. XVII. Formation mechanisms underlying natur-
ally occurring DNA deletions in the human genome and their potential
relevance for bridging the gap between induced DNA double-strand
breaks and deletions in irradiated germ cells. Mutat Res. 753:114–130.
Savage JR. 1993. Interchange and intra-nuclear architecture. Environ Mol
Mutagen. 22:234–244.
Sax K. 1940. An analysis of X-ray induced chromosomal aberrations in
Tradescantia. Genetics. 25:41–68.
Schipler A, Iliakis G. 2013. DNA double-strand-break complexity levels and
their possible contributions to the probability for error-prone process-
ing and repair pathway choice. Nucleic Acids Res. 41:7589–7605.
Singh SK, Hawkins C, Clarke ID, Squire JA, Bayani J, Hide T, Henkelman
RM, Cusimano MD, Dirks PB. 2004. Identification of human brain
tumour initiating cells. Nature. 432:396–401.
Sutherland BM, Bennett PV, Sidorkina O, Laval J. 2000. Clustered DNA
damages induced in isolated DNA and in human cells by low doses of
ionizing radiation. Proc Natl Acad Sci USA. 97:103–108.
Takata H, Hanafusa T, Mori T, Shimura M, Iida Y, Ishikawa K, Yoshikawa K,
Yoshikawa Y, Maeshima K. 2013. Chromatin compaction protects gen-
omic DNA from radiation damage. PLoS One. 8:e75622.
Thacker J, Wilkinson RE, Goodhead DT. 1986. The induction of chromo-
some exchange aberrations by carbon ultrasoft X-rays in V79 hamster
cells. Int J Radiat Biol Relat Stud Phys Chem Med. 49:645–656.
Thompson LH. 2012. Recognition, signaling, and repair of DNA double-
strand breaks produced by ionizing radiation in mammalian cells: the
molecular choreography. Mutat Res. 751:158–246.
Vakifahmetoglu H, Olsson M, Zhivotovsky B. 2008. Death through a tra-
gedy: Mitotic catastrophe. Cell Death Differ. 15:1153–1162.
Vasireddy RS, Karagiannis TC, El-Osta A. 2010. gamma-radiation-induced
gammaH2AX formation occurs preferentially in actively transcribing
euchromatic loci. Cell Mol Life Sci. 67:291–294.
Wiklund K, Fernandez-Varea JM, Lind BK. 2011. A Monte Carlo program
for the analysis of low-energy electron tracks in liquid water. Phys
Med Biol. 56:1985–2003.
Williams D. 2008. Radiation carcinogenesis: Lessons from Chernobyl.
Oncogene. 27(Suppl. 2):S9–18.
Woodbine L, Gennery AR, Jeggo PA. 2014. The clinical impact of defi-
ciency in DNA non-homologous end-joining. DNA Repair (Amst).
16:84–96.
Xu Y, Price BD. 2011. Chromatin dynamics and the repair of DNA double
strand breaks. Cell Cycle. 10:261–267.
Yokota H, Van Den Engh G, Hearst JE, Sachs RK, Trask BJ. 1995. Evidence
for the organization of chromatin in megabase pair-sized loops
arranged along a random walk path in the human G0/G1 interphase
nucleus. J Cell Biol. 130:1239–1249.
Zhang Y, Mccord RP, Ho YJ, Lajoie BR, Hildebrand DG, Simon AC,
Becker MS, Alt FW, Dekker J. 2012. Spatial organization of the mouse
genome and its role in recurrent chromosomal translocations. Cell.
148:908–921.
12
D.A. FRIEDMAN ET AL.
Downloaded by [University of California, San Diego] at 06:07 14 March 2016


---
*Extraction method: pymupdf*