Somatic Expansion in Mouse and Human Carriers
of Fragile X Premutation Alleles
Rachel Adihe Lokanga,1Ali Entezam,1†Daman Kumari,1Dmitry Yudkin,1Mei Qin,2Carolyn Beebe Smith,2and Karen Usdin1∗
1Section on Gene Structure and Disease, National Institute of Diabetes, Digestive and Kidney Diseases, National Institutes of Health,
Bethesda, Maryland;2Section on Neuroadaptation and Protein Metabolism National Institute of Mental Health, National Institutes of Health,
Communicated by Haig H. Kazazian, Jr.
Received 24 April 2012; accepted revised manuscript 17 July 2012.
Published online 8 August 2012 in Wiley Online Library (www.wiley.com/humanmutation). DOI: 10.1002/humu.22177
pansion of a specific tandem repeat. The three fragile X-
related disorders (FXDs) arise from germline expansions
of a CGG•CCG repeat tract in the 5?UTR (untranslated
region) of the fragile X mental retardation 1 (FMR1)
gene. We show here that in addition to germline expan-
sion, expansion also occurs in the somatic cells of both
mice and humans carriers of premutation alleles. Expan-
sion in mice primarily affects brain, testis, and liver with
very little expansion in heart or blood. Our data would be
consistent with a simple two-factor model for the organ
specificity. Somatic expansion in humans may contribute
to the mosaicism often seen in individuals with one of
the FXDs. Because expansion risk and disease severity
are related to repeat number, somatic expansion may ex-
acerbate disease severity and contribute to the age-related
increased risk of expansion seen on paternal transmission
in humans. As little somatic expansion occurs in murine
lymphocytes, our data also raise the possibility that there
may be discordance in humans between repeat numbers
measured in blood and that present in brain. This could
explain, at least in part, the variable penetrance seen in
some of these disorders.
Hum Mutat 34:157–166, 2013. Published 2012 Wiley Periodi-
Repeat expansion diseases result from ex-
KEY WORDS: fragile X-related disorders; FMR1; somatic
The repeat expansion diseases (REDs) are a group of human ge-
netic conditions that result from expansion of a particular tandem
Additional Supporting Information may be found in the online version of this article.
†Present address: GeneDx, Gaithersburg, Maryland.
∗Correspondence to: Karen Usdin, Building 8, Room 2A19, National Institutes of
Health, 8 CENTER DR MSC 0830, Bethesda, MD 20892–0830. E-mail: email@example.com
Contract grant sponsors: Intramural program of the NIDDK, NIH (DK057808 to K.U.).
MIM# 309550). Fragile X (FX)-associated tremor and ataxia syn-
ovarian insufficiency (FXPOI) [Allingham-Hawkins et al., 1999;
Conway et al., 1998; Murray et al., 1998; Vianna-Morgante et al.,
1996] result from inheritance of FMR1 alleles with 55–200 repeats.
Such alleles are referred to as FX premutation alleles (PM). FX-
TAS is a neurodegenerative disorder that results in gait and balance
abnormalities, cognitive decline, dementia, and executive function
deficits. FXPOI symptoms include menstrual irregularities, fertility
problems, and an early menopause. FXPOI is thought to account
et al., 1999]. Evidence suggests that the pathology in FXTAS, and
the elevated expression of RNA-containing long CGG repeat tracts
[Handa et al., 2005; Hashem et al., 2009; Jin et al., 2003].
tional transfer. Paternal transmissions of such alleles generate small
increases in repeat number with an effect of paternal age being evi-
can have two consequences. Because PM pathology is related to re-
peat number [Tassone et al., 2007], this could result in a gradual
increase in disease severity in families with the PM. Furthermore,
because repeat number is related to the risk of further expansion,
it could also affect expansion risk on subsequent intergenerational
transmissions. Maternal transmissions of PM alleles generate large
increases in repeat number. When the repeat number exceeds 200,
a third disorder, FX syndrome (FXS), the most common heritable
cause of intellectual disability, is seen in those who inherit such
The REDs-associated repeats form a variety of secondary struc-
tures including hairpins, tetraplexes, and triplexes (reviewed in
[Mirkin, 2007]). It is generally thought that expansions arise due to
REDs-associated repeats in human cell extracts, human cell lines,
bacteria, and yeast suggest that any one of a variety of biological
processes including replication, transcription, DNA repair, and re-
have previously shown that the ATM (ataxia-telangiectasia, mu-
tated) and the ATR (ATM and Rad3-related) DNA damage re-
sponse pathways are involved in reducing the risk of expansion
during intergenerational transfer [Entezam and Usdin, 2008, 2009]
and that factors such as oxidative stress can exacerbate expansion
risk [Entezam et al., 2010]. Although the molecular details of the
Published 2013 Wiley Periodicals, Inc.∗This article is a US Government work and, as such, is in the public domain of the United States of America.
mechanism responsible for repeat expansions remain unclear, these
sions occur in the gamete by a process that likely involves aberrant
DNArepair.However, whether allREDssharethesamemechanism
for intergenerational expansions is unclear.
Somatic expansions are seen in many of the other REDs [Chong
et al., 1995; De Biase et al., 2007; Manley et al., 1999a; Tanaka et al.,
1999; Telenius et al., 1994; Ueno et al., 1995]. Evidence from mouse
models of these diseases suggests that such expansions can exacer-
bate disease severity and reduce the age at which disease symptoms
become apparent [Wheeler et al., 2003]. Whether somatic expan-
sion involves the same mechanism as intergenerational expansion
is unknown. Somatic expansion of FX PM alleles has not yet been
reported, although mosaicism with respect to CGG•CCG repeat
many of the heterogeneous products seen on Southern blot analysis
of FM alleles represent an artifact of electrophoresis in the presence
of ethidium bromide, more than one allele size is frequently seen
even when analysis is done in the absence of this intercalator [Nolin
et al., 2008]. A striking example has been reported of an individual
who has unmethylated repeats ranging in size from the PM to the
FM range in most parts of the brain but a single methylated FM
allele in the parietal lobe and most nonbrain tissues studied [Tay-
lor, et al., 1999]. However, without knowing the size of the original
by somatic expansion of smaller ones. Because the consequences
of repeat expansion are thought to result from the combination of
FMR1 mRNA levels and the number of repeats [Jin et al., 2003],
true somatic expansion, should it occur in FX PM carriers, would
have the potential to produce a transcript that would be more dele-
terious than the original allele. It could also lead to an increased
risk of transmission of larger expansions with increasing parental
model for the FX PM which recapitulates features of both the
neurodegeneration [Entezam et al., 2007] and ovarian dysfunction
[Hoffman et al., 2012] seen in human PM carriers as well as the in-
tergenerational repeat expansion seen in this population [Entezam
et al., 2007, 2010; Entezam and Usdin, 2008, 2009]. We show here
that somatic expansion is also seen in these mice. Some organs are
in a lymphoblastoid cell line from a human PM carrier and in the
brains of FXTAS patients. Our data may have implications not only
for the understanding of the expansion mechanism but also for the
assessment of expansion risk in families with PM alleles.
Materials and Methods
Mouse and Human Samples
The generation of the FX PM mice was described previously
[Entezam et al., 2007]. Mice were maintained in accordance with
the guidelines of the NIDDK Animal Care and Use Committee and
with the Guide for the Care and Use of Laboratory Animals (NIH
publication number 85-23, revised 1996). Sperm was isolated from
mouse epididymis using standard procedures [Nagy et al., 2003].
Somatic and germ line cells were isolated from testes as previously
brain were isolated as follows. Mice were anesthetized with sodium
pentobarbital (100 mg/kg, i.p.), decapitated, and the brains frozen
in dry ice. The cerebellum was dissected and different brain regions
were punched by means of Harris Uni-Core (1.25 mm) (Electron
Microscopy Sciences, Hatfield, PA) in a Leica cryostat set at–22◦C.
Human tissue samples from individuals with FXTAS were ob-
tained from the NICHD Brain and Tissue Bank for Developmental
Disorders (Baltimore, MD). DNA isolated from the blood of one
of these individuals was a kind gift from Sally Nolin (Institute for
tute for Medical Research (Camden, NJ) and maintained in culture
usingRPMI1640 withGlutaMax,10%heatinactivated fetal bovine
serum and 1X antibiotics (Life Technologies, Grand Island, NY)
at 37◦C in a 5% CO2atmosphere. The identity of the original cell
line and subsequent passages showing expansion was confirmed by
SNP analysis of 26 loci using the Sequenom MassARRAY iPLEXTM
platform (Bioserve Biotechnologies, Beltsville, MD).
Quantitation of mRNA
Total RNA from different mouse organs was isolated using
TrizolTM(Life Technologies) as per themanufacturer’s instructions.
The RNA was then treated with DNase I and reverse transcribed
using a SuperScript
gies). Real-time PCR was done in triplicate using TaqMan
universal PCR master mix and the appropriate Taqman probe–
primer pairs (Applied Biosystems). Because the expression levels
of the endogenous controls tested (β-actin, Gapdh, and Ubc9) dif-
fered in various organs, they could not be used for normalization.
However, equal amounts of RNA were used in these experiments
and so normalization was carried out by comparing the Ct value
for the mRNA in different organs to the Ct value obtained for
R ?VILOTMcDNA synthesis kit (Life Technolo-
Analysis of Repeat Length
Genomic DNA was prepared from mouse and human samples as
previously described [Entezam et al., 2007]. The size of the mouse
CGG•CCG repeat tract was determined as described previously
[Lavedan et al., 1998] except that primer frax-m4 was either labeled
with 6-carboxyfluorescein (FAM) or 4,7,2?,4?,5?,7?-hexachloro-6-
carboxyfluorescein (HEX). The 5?and 3?primers are 25 and 24
AF (5?-GCTGCCAGGGGGCGTGCGGCA-3?) primers were used.
This primer pair produces a PCR product that contains the repeat
and 77 bp of DNA from the upstream flank and 41 bp from the
downstream flank for a total of 124 bp of flanking DNA. The so-
from each age group [Lee et al., 2010].
Determination of Msh2, Msh3, and Msh6 Protein Levels
in Different Mouse Organs
Organs were collected from 2-month-old PM mice, the pro-
teins isolated, resolved by SDS-PAGE and Western blotted using
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
standard procedures. Msh2, Msh3, and Msh6 were detected using
the antibodies AB70270 (Abcam, Cambridge, MA), AB74607 (Ab-
cam), and 610918 (BD Biosciences, Sparks, MD), respectively, and
an ECL prime kit (GE Healthcare, Piscataway, NJ) used according
to the suppliers’ instructions. Purified human MSH2/MSH3 used
as a positive control was a kind gift of Shikha Gupta (NIH).
A Unique Pattern of Organ-Specific Somatic Expansion
Is Seen in FX PM Mice
To assess whether somatic instability occurs in FX PM mice, we
on a 3130XL Genetic Analyzer and analyzed using GeneMapper
4.0 software. Typically, the resultant GeneMapper profiles have a
Gaussian distribution of products that differ from each other by
one repeat. Products larger and smaller than the mean allele size
are sometimes referred to as “stutter-products.” They thought to be
due, in part, to strand-slippage during PCR. However, work with
mutants that abolish expansion suggests this explanation accounts
for the smaller products but not all of the larger products [Lokanga
et al. (In preparation)]. In samples containing a mixture of alleles
with different repeat numbers, as would be seen if expansions or
deletions had occurred in the tissue from which the sample was
derived, the product distribution is broader, right-shifted or left-
shifted in the case of expansions and deletions, respectively, and
may contain multiple peaks as new derivative alleles arise in the cell
The number of CGG•CCG repeats in the DNA from different
PCR product, and compared to the repeat number found in the
tail DNA of the same animals at 3 weeks of age. When DNA from
the organs of a number of PM mice was tested at 2 months of
age, not only was the repeat number no different from the repeat
GeneMapper profiles had the same relatively narrow distribution
of PCR products (see Fig. 1 for a representative example). This
indicates that little or no somatic instability occurs during early
postnatal life in FX PM mice. There was one notable exception.
Testes DNA showed a small right-shift of the PCR products, with
the most abundant product now having a mobility consistent with
the addition of one repeat unit. A low level of somatic expansion
and/or skewing of the GeneMapper peak profile in the case of brain
and liver or by a rightward shift in the peak distribution toward
alleles with ∼2 additional repeats in testes (Fig. 1, middle of panel
A). By 12 months of age, expansions were much more extensive in
all animals tested, with clear evidence of expansion seen in a variety
of organs particularly brain, liver, and testes (Fig. 1, right-hand side
of panel A). Unlike what has been reported for mouse models of
other REDs, kidney showed relatively little expansion. Heart and
blood showed even less expansion, even in mice that were more
than 24 months of age (Fig. 1, panels A and B and data not shown).
with similar numbers of repeats. The change in the average somatic
instabilityindexwith age for micewith 141–145 repeats isshown in
Different Expansion Profiles Are Apparent in Different
In liver and testes the allele profile frequently showed two ma-
jor allele sizes, one corresponding to the original allele and one
that was 7–14 repeats larger. In total brain, a pattern was seen that
was consistent with the addition of a smaller number of repeats.
Different regions within the brain also showed different expansion
profiles with thalamus showing relatively little expansion, whereas
the striatum and basolateral amygdala showed more expansions
with a “jump size” that was 6 and 8 repeats, respectively (Fig. 2).
High Levels of Expansion Are Seen in the Nonsomatic
Cells of the Testis
The repeat profile in sperm isolated from 12-month-old animals
differs from that seen in total testes DNA from the same animal
(Fig. 3A). In sperm, a single peak was seen that was five repeats
larger than the original allele, whereas in testes two peaks are seen,
one corresponding to the original sized allele and one correspond-
ing to an allele that was 13 repeats larger. Small pool (SP) PCR
shows that expansion occurs in ∼75% of sperm with deletions be-
ing seen at a frequency of ∼10% (Fig. 3B and data not shown).
The expansion frequency and average expansion size in sperm is
very similar to that seen in the offspring of males with the same
repeat number [Entezam et al., 2007, 2010; Entezam and Usdin,
the PM allele probably occur predominantly prezygotically consis-
tent with our previous observations [Entezam et al., 2010; Entezam
and Usdin, 2008, 2009]. Within the testis of a mouse that was 11
months old, expansions are apparent in both the fraction of cells
enriched for primary spermatocytes and the fraction enriched for
spermatogonial stem cells (Fig. 3C). However, no expansions were
in either somatic or germ cells of prepubertal (18 days old) mice
(Fig. 3D), suggesting that the expansions do not arise in the pri-
mordial germ cells or prospermatogonia, but presumably begin to
accumulate in germ cells during the peripubertal period as sper-
Somatic Expansion Is not Related to Proliferative Capacity
It has been suggested that expansion is the result of a problem
arising during DNA replication. Liver is an organ that has one of
the highest levels of expansion, with this first becoming apparent
in adult mice. Yet, very little cell division or DNA synthesis occurs
in the livers of adult rodents [Bohman et al., 1985]. Similarly brain,
which contains a high proportion of slow-growing glia and non-
dividing neurons, shows extensive expansion. The amount of ex-
pansion differed in different regions of the brain with the striatum,
basolateral amygdala, and cerebellum showing larger expansions
than the frontal cortex or thalamus (Fig. 2). In some regions of the
brains of some mice, expansion was so extensive that most alleles
present in the population were larger than the original allele (e.g.,
in the cerebellum shown in Fig. 2).
FMR1 Transcription Levels Do not Explain the Tissue
Specificity of Expansion
Transcription of the gene in which the repeat is located has also
been suggested to be a major factor in the expansion process. We
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
organs of male mice of different ages. Representative results for mice at three different ages are shown. The gray dotted vertical line in each
panel indicates the position of the original-sized allele based on the repeat number in the tail DNA taken at weaning (3 weeks of age; tail 1). The
second tail DNA sample in each case (tail 2) was taken after euthanasia. The repeat DNA profiles of the 12-month-old animal was generated using
an HEX-labeled primer. The repeat profiles for the 2- and 4-month-old animals were generated using an FAM-labeled primer. The choice of primer
molecular weight marker was used for the 12-month-old animal. B: GeneMapper profiles from blood and heart of a 12-month-old animal. C: Graph
showing the average somatic instability index for mice with 141–145 repeats at 2, 4, and 12 months.
Age and organ dependent somatic expansions in FX PM mice. A: GeneMapper profiles were obtained for the repeat tract in different
thusmeasuredFmr1mRNAlevels intheorgans oftheFXPMmice.
explain the low rate of expansion in this organ, Fmr1 mRNA levels
in kidney, which shows very little expansion, are much higher than
they are in liver, which shows high levels of expansion. Hence, there
is no simple relationship between the levels of Fmr1 transcription
and somatic expansion frequency.
Levels of the Mismatch Repair Proteins Msh2, Msh3, and
Msh6 also Do not Explain the Tissue Specificity
The mismatch repair proteins Msh2 and Msh3 form a complex,
MutSβ, which is required for expansion in mouse models of some
of the other REDs [Foiry et al., 2006; Manley et al., 1999b; Tome
et al., 2009; Wheeler et al., 2003] and which our data suggests is
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
The dotted line indicates the mobility of the original allele as assessed from the tail DNA taken at weaning. GeneMapper profiles for the FX repeat
were obtained from different regions of the brain of a 12-month-old PM male mouse with ∼131 repeats. Similar profiles were seen in two other
mice that were tested.
also required for germline and somatic expansion in the FX PM
mice (Lokanga and Usdin, manuscript in preparation). Because the
mRNA levels for Msh2, Msh3, and the alternate binding partner
of Msh2 and Msh6 do not correlate well with the levels of their
respective proteins (cf., Fig. 4A with Supp. Fig. S1), we compared
the levels of these proteins with the tendency of the heart, brain,
liver, kidney, and testes to show expansion. Because Msh3 levels are
thought to be rate limiting for the formation of the MutSβ complex
Msh3 is rapidly degraded, the level of Msh3 in any organ is thought
to be a good indicator of the amount of MutSβ present. Because
brain and testes—organs that have high levels of expansion—also
have high levels of Msh3, this would be consistent with the idea
that the levels of Msh3/MutSβ determine the propensity of a given
organ to undergo expansion. However, the fact that heart, kidney,
and liver all have similar levels of Msh3, would not be consistent
with that hypothesis because all of these organs show very different
levels of somatic expansion.
Expansion from a PM Allele to an FM Allele Can Be Seen
in a Human Lymphoblastoid Cell Line
The somatic expansion seen in mice led us to examine human
DNA samples in our collection that had been isolated from a lym-
phoblastoid cell line, GM06891, over a period of 42 months. This
cell line originally had a single predominant allele of ∼118 repeats
(t0). We found that in samples isolated after two years of inter-
mittent propagation, very little of the original allele was detectable
and alleles that were ∼16, ∼35, and 51 repeats larger were apparent
with the largest allele predominating (t1 in Fig. 5). A later sample
(t2) showed that the most prominent allele in the sample had ∼195
repeats, that is, it was 77 repeats larger than the original allele. In
addition, there was a small amount of alleles that were ∼45, 62,
and 93–96 repeats larger than the original allele. As the PCR yield
ments is not surprising. Nonetheless, these results were verified by
Southern blotting and a different PCR technique by the Institute
for Basic Research on Developmental Disabilities (Staten Island,
NY). The results, which confirm our findings, are shown in Supp.
The identity of the original cell line and subsequent passages
showing expansion was confirmed by SNP analysis using 26 loci.
The chance that the cell populations containing expansions were
not derived from the original cell line is 1.042 × 10–11. As selec-
tion is thought to operate in favor of shorter alleles that are likely to
we observed likely represent bona fide somatic expansions. As with
some mouse tissues, expansion in this cell line also seemed to occur
in a saltatory fashion with a jump size, in this instance, of 15–19
repeats. In this case, the cumulative expansions resulted in the con-
these expansions arise and to identify the factors that led to their
Somatic Expansions Are also Seen in Human Brain
The in vitro expansion of the FX repeats in a human cell line
raised the possibility that somatic expansions may also occur in
humans in vivo. To test this idea, we examined autopsy samples
from three individuals with FXTAS. These individuals had repeat
numbers in blood of 67, 82, and 91 at diagnosis. The 91 repeat
allele had one AGG interruption at the 5?end of the allele (S. Nolin,
personal communication, 2012). The interspersion pattern of the
from the patient with 67 repeats: the prefrontal cortex and the
cerebellum. For the patient with the 67 repeat allele, very similar
GeneMapper profiles were obtained for both regions with a narrow
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
based on the tail DNA at 3 weeks of age in the case of the data shown in Panels A, B, and C and 18 days in the case of the data shown in Panel
D. The repeat profile in Panel A was obtained with an HEX-labeled primer and the repeat profiles shown in B, C, and D were obtained using an
FAM-labeled primer. The choice of primer label does not affect the repeat profile. The numbers in panel A, B, and C refer to the number of repeats
added to the major allele in the population relative to the repeats present in the original allele. A: Comparison of the repeat number in tail DNA
taken at weaning and in totaltestes DNAand sperm DNAfrom a 12-month-old animal. B: Comparison of the repeat number in tailtaken at weaning
with the repeat number in the products of 3 independent small-pool PCRs (labeled a, b, and c) carried out on the sperm of the same animal at 12
months of age. C, D: Comparison of the repeat profile in the tail DNA taken at weaning and in the fractions enriched for the indicated cell type in
the testes of mice at 11 months (C) and 18 days (D).
Cell-specific and age-related expansion in the testes. The dotted line in each panel represents the original allele size (∼140 repeats)
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
The average mRNA levels from three animals was calculated and the
Fmr1transcript levels inbrain, kidney,liver,and testisnormalized tothe
levels in heart. B: Levels of Msh2, Msh3, and Msh6 proteins in different
organs. Representative data from one mouse is shown—B: brain, H:
heart; L: liver; K: kidney; and T: testes. Similar levels of proteins were
seen in two other animals tested.
Fmr1 mRNA and Msh2, Msh3 and Msh6 protein levels in
distribution of PCR products consistent with little or no instability
(data not shown). In the case of the individual with 82 repeats,
tissue from heart was available. Because mouse heart shows little,
if any, instability we rationalized that the human heart sample may
in the repeat profile were seen in this individual (Supp. Fig. S3).
No heart DNA was available from the individual with 91 repeats.
individual 6 years prior to death. Although no evidence of a change
line. GeneMapper profiles for the PM repeat obtained for the GM06891
the same amount of LIZ1200 standard was used. However, because of
the poor PCR yield of the sample at t2, a phenomenon typical of ampli-
fication through long repeat tracts, this data is shown on an expanded
inal allele and the green and blue profiles corresponding to samples t1
and t2. The data in this panel are all shown on the same scale. The dot-
tedlineindicates thesizeoftheoriginal allele.Thenumbers associated
allele at t0.
Somatic expansions in a human PM lymphoblastoid cell
blood (Fig. 6). The frontal cortex showed a slight shift in the major
peak corresponding to an increase of one repeat. There was also a
small peak corresponding to an allele that was ∼8 repeats larger,
and a smaller peak just discernable above background that was ∼19
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
regions of this patient compared to the profile seen in DNA isolated from blood isolated at the time of diagnosis six years before death. B: Repeat
profiles for different brain regions superimposed with the repeat profile from heart shown in black. The numbers refer to the number of repeats
added relative to the major allele in blood. This profile was reproducible and all peaks were seen with a range of different DNA concentrations.
Expansions differ in different regions of the brain of an FXTAS patient with 91 repeats. A: GeneMapper profiles from different brain
repeats larger than the allele seen in blood. Whether the expansions
occur uniformly in these brain regions or occur preferentially in
specific cell populations or subregions of the cortex and amygdala
remains to be seen.
We have shown here that FX PM mice show age-dependent so-
matic expansion of the CGG•CCG repeat that occurs primarily in
in men with FXTAS [Hessl et al., 2007]. Expansion also occurs in
human PM cells in tissue culture converting the PM allele stepwise
into an FM allele (Fig. 5). Expansion was also detected in the brains
of individuals with FXTAS (Fig. 6 and Supp. Fig. S3). Although the
of interruptions in the largest allele is taken into consideration, it
should be noted that the largest human allele examined was ∼50
repeats shorter than the mouse alleles studied here. Because the re-
peat number in the starting allele has a significant effect on both
the size and frequency of expansion in the germline [Entezam et al.,
2007], it may well be that it also affects the frequency and number
of repeats added during somatic expansion. Furthermore, we know
that in mice both genetic and environmental factors can exacerbate
expansion risk [Entezam et al., 2010; Entezam and Usdin, 2008,
2009]. Thus, depending on repeat size, genetic background, and
environmental exposure, somatic expansion may be more extensive
in some human PM carriers than others. Our demonstration that
human lymphoblastoid cells with an initial starting allele with 118
repeats show significant expansion suggests that somatic expansion
may be particularly significant in that subset of PM carriers that
have >100 repeats. The fact that little, if any, expansion was seen
in the blood of PM mice, also raises the possibility that in humans
there may also be discordance between the repeat number seen in
blood in these individuals and that seen in more expansion-prone
regions such as the brain, liver, and gonads.
The high level of expansion in the brain and liver of the FX PM
mice, organs in which most cells are postmitotic, suggests that cell
in FX PM mice may be more likely to result from aberrant DNA
damage repair than replicative DNA synthesis. In this respect, so-
transmission of PM alleles where expansion is seen in oocytes and
is exacerbated by oxidative damage [Entezam et al., 2010; Entezam
and Usdin, 2008, 2009].
Somatic expansion is also seen in mouse models of other REDs.
of somatic expansion are seen in the kidney [Fortune et al., 2000;
Gomes-Pereira et al., 2001; Lia et al., 1998; Mangiarini et al., 1997;
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
relatively little propensity to expand (Fig. 1). In mouse models of
Friedreich ataxia, FRDA, a GAA•TTC-RED, very little expansion is
mice do not. Understanding the reasons for the differences in tissue
specificity may help us identify some of the factors important for
driving expansions in different disease models and thus ultimately
shed light on the expansion mechanism.
Avarietyofdifferentexplanations have beenadvanced toexplain
the tissue specificity of expansion in other REDs. Transcription of
the affected gene has been suggested to be important for expansion
[Lin et al., 2009, 2010; Lin and Wilson, 2007; McIvor et al., 2010;
Mochmann and Wells, 2004; Nakamori et al., 2011; Schumacher
et al., 2001]. However, the fact that heart, which shows almost no
expansion, and liver and testes, which show extensive expansion
have similar levels of Fmr1 mRNA (Fig. 4A), suggests that Fmr1
mRNA transcription alone does not explain the organ specificity of
expansion in the FX PM mice.
Variations in the levels of different proteins involved in DNA
replication and/or repair have also been suggested to account for
the tissue specificity of repeat expansion. For example, differences
in the levels of the Flap endonuclease, FEN1, and Polβ—enzymes
involved base excision repair—have been suggested to account for
the difference in expansions in cerebellum and striatum in a mouse
model of HD [Goula et al., 2009]. However, direct evidence for a
role of these proteins in the REDs has not yet been demonstrated
and the level of these proteins was not examined in other tissues
[Goula et al., 2009]. High levels of MutSβ have been suggested to
account for the fact that expansions seen in induced pluripotent
stemcells (iPSCs) derived frompatients with FRDA are much more
extensive than those seen in the fibroblasts from which they were
derived [Ku et al., 2010; Seriola et al., 2011]. Similarly, expansion is
higher in embryonic stem cells (ESCs) derived from embryos with
myotonic dystrophy type 1 (DM1), a CTG•CAG-RED, than it is in
differentiated cells produced from these ESCs [Seriola et al., 2011].
However, a larger study comparing multiple tissues concluded that
the levels of these proteins did not account for the tissue specificity
of expansion in mouse model for another CTG•CAG-RED, Hunt-
ington’s disease (HD) [Lee et al., 2010]. One caveat with this data
is that its conclusion was based on the Msh2/3 mRNA levels, which
we have shown to not correlate well with the Msh2/3 protein levels
in the FX PM mice (Fig. 4B and Supp. Fig. S1). The fact that, in
the FX PM mice, the levels of these proteins are very high in both
brain and testis would be consistent with a role for MutSβ levels in
determining organ specificity. However, very similar levels of Msh3
are seen in heart, kidney, and liver despite the fact that these organs
have very different propensities to expand. Thus, when protein lev-
els are considered and multiple tissues are compared there is not
a good correlation between the levels of MutSβ and the expansion
However, it is unnecessary to invoke complex models to explain
our data as has been done for other REDs [Lee et al., 2010]. Our
observation that expansion in the FX PM mice likely occurs via a
process involving aberrant DNA repair raises the possibility that
some combination of the levels of MutSβ and the amount of DNA
damage may account for the organ specificity. The amount of DNA
damage, as assessed by the number of γ-H2AX foci, does not alone
account for the organ specificity because brain and heart have very
similar numbers of foci [Hudson et al., 2011; Wang et al., 2009],
despitetheir very different levels ofexpansion. However, when con-
organs can be reconciled. For example, expansions would be low in
tissue such as heart where MutSβ and DNA damage levels are low.
Expansions would be high in brain and testes because the levels of
expansion will occur in that organ. The prediction from this model
would be that agents that increase the amount of DNA damage in
organs that normally show little somatic instability would produce
increased levels ofexpansion in those organs. Work is in progress to
test this hypothesis.
Although the details of the mechanism responsible for somatic
expansion of FX PM alleles remains unknown, our demonstration
expected to increase the deleterious effect of transcripts produced
from such alleles. The intertissue variability in expansion risk also
are genetic or environmental factors that affect somatic expansion
risk. In addition, expansion in testes has the potential to affect the
size of the repeat transmitted by a PM father to his daughter. This
could account for the observed effect of paternal age on the size of
the transmitted PM allele in humans [Ashley-Koch et al., 1998]. We
have previously shown that environmental factors such as oxidative
stress can increase intergenerational expansion risk [Entezam et al.,
2010]. Identifying factors that reduce somatic expansion may lead
to the development of strategies to reduce disease severity in PM
carriers and the risk of parental transmissions of expanded alleles.
We would like to thank Huiyan Liu (NIDDK) for invaluable assistance and
mental Disabilities), for their testing of some of our human DNA samples.
We are also grateful for the hard-working technicians who take care of our
mouse colony without whose efforts this work would not be possible.
Allingham-Hawkins DJ, Babul-Hirji R, Chitayat D, Holden JJ, Yang KT, Lee C, Hud-
son R, Gorwill H, Nolin SL, Glicksman A, Jenkins EC, Brown WT, et al. 1999.
Fragile X premutation is a significant risk factorforpremature ovarian failure: the
International Collaborative POF in Fragile X study—preliminary data. Am J Med
Ashley-Koch AE, Robinson H, Glicksman AE, Nolin SL, Schwartz CE, Brown WT,
Turner G, Sherman SL. 1998. Examination of factors associated with instability of
the FMR1 CGG repeat. Am J Hum Genet 63:776–785.
BohmanR,Tamura CT,Doolittle MH,CascaranoJ.1985.Growth andagingintherat:
changes in total protein, cellularity, and polyploidy in various organs. J Exp Zool
of the human DNA mismatch repair system. J Biol Chem 275:18424–18431.
Chong SS, McCall AE, Cota J, Subramony SH, Orr HT, Hughes MR, Zoghbi HY. 1995.
Gametic and somatic tissue-specific heterogeneity of the expanded SCA1 CAG
repeat in spinocerebellar ataxia type 1. Nat Genet 10:344–350.
Clark RM, De Biase I, Malykhina AP, Al-Mahdawi S, Pook M, Bidichandani SI. 2007.
The GAA triplet-repeat is unstable in the context of the human FXN locus and
displays age-dependent expansions in cerebellum andDRG in a transgenic mouse
model. Hum Genet 120:633–640.
Conway GS, Payne NN, Webb J, Murray A, Jacobs PA. 1998. Fragile X premutation
screening in women with premature ovarian failure. Hum Reprod 13:1184–1187.
De Biase I, Rasmussen A, Endres D, Al-Mahdawi S, Monticelli A, Cocozza S, Pook
M, Bidichandani SI. 2007. Progressive GAA expansions in dorsal root ganglia of
Friedreich’s ataxia patients. Ann Neurol 61:55–60.
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
Entezam A, Biacsi R, Orrison B, Saha T, Hoffman GE, Grabczyk E, Nussbaum RL, Download full-text
Usdin K. 2007. Regional FMRP deficits and large repeat expansions into the full
mutation range in a new fragile X premutation mouse model. Gene 395:125–134.
Entezam A, Lokanga AR, Le W, Hoffman G, Usdin K. 2010. Potassium bromate, a
potent DNA oxidizing agent, exacerbates germline repeat expansion in a fragile X
premutation mouse model. Hum Mutat 31:611–616.
Entezam A, Usdin K. 2008. ATR protects the genome against CGG.CCG-repeat expan-
sion in fragile X premutation mice. Nucleic Acids Res 36:1050–1056.
Entezam A, Usdin K. 2009. ATM and ATR protect the genome against two different
types of tandem repeat instability in fragile X premutation mice. Nucleic Acids
Foiry L, Dong L, Savouret C, Hubert L, te Riele H, Junien C, Gourdon G. 2006. Msh3
is a limiting factor in the formation of intergenerational CTG expansions in DM1
transgenic mice. Hum Genet 119:520–526.
Fortune MT, Vassilopoulos C, Coolbaugh MI, Siciliano MJ, Monckton DG. 2000.
Genschel J, Littman SJ, Drummond JT, Modrich P. 1998. Isolation of MutSbeta from
human cells and comparison of the mismatch repair specificities of MutSbeta and
MutSalpha. J Biol Chem 273:19895–19901.
Gomes-Pereira M, Fortune MT, Monckton DG. 2001. Mouse tissue culture models of
unstable triplet repeats: in vitro selection for larger alleles, mutational expansion
bias and tissue specificity, but no association with cell division rates. Hum Mol
Goula AV, Berquist BR, Wilson DM, 3rd, Wheeler VC, Trottier Y, Merienne K. 2009.
Stoichiometry of base excision repair proteins correlates with increased somatic
CAG instability in striatum over cerebellum in Huntington’s disease transgenic
mice. PLoS Genet 5:e1000749.
Hagerman RJ, Leehey M, Heinrichs W, Tassone F, Wilson R, Hills J, Grigsby J, Gage
B, Hagerman PJ. 2001. Intention tremor, parkinsonism, and generalized brain
atrophy in male carriers of fragile X. Neurology 57:127–130.
Handa V, Goldwater D, Stiles D, Cam M, Poy G, Kumari D, Usdin K. 2005. Long
CGG-repeat tracts are toxic to human cells: Implications for carriers of Fragile X
premutation alleles. FEBS Lett 579:2702–2708.
Ectopic expression of CGG containing mRNA is neurotoxic in mammals. Hum
Mol Genet 18:2443–2451.
RJ. 2007. Amygdala dysfunction in men with the fragile X premutation. Brain
Hoffman GE, Le W, Entezam A, Otsuka N, Tong Z-B, Nelson L, Flaws JA, McDonald
JH, Jafar S, Usdin K. 2012. Ovarian abnormalities in a mouse model of Fragile
X-associated primary ovarian insufficiency. J Histochem Cytochem 60:439–456.
and persistence of radiation-induced DNA damage is more pronounced in young
animals than in old animals. Aging (Albany NY) 3:609–620.
Jin P, Zarnescu DC, Zhang F, Pearson CE, Lucchesi JC, Moses K, Warren ST. 2003.
RNA-mediated neurodegeneration caused by the fragile X premutation rCGG
repeats in Drosophila. Neuron 39:739–747.
Ku S, Soragni E, Campau E, Thomas EA, Altun G, Laurent LC, Loring JF, Napierala
M, Gottesfeld JM. 2010. Friedreich’s ataxia induced pluripotent stem cells model
intergenerational GAATTC triplet repeat instability. Cell Stem Cell 7:631–637.
within the first exon of the human FMR1 gene are not intrinsically unstable in
transgenic mice. Genomics 50:229–240.
Boily MJ, et al. 2010. A novel approach to investigate tissue-specific trinucleotide
repeat instability. BMC Syst Biol 4:29.
Lia AS, Seznec H, Hofmann-Radvanyi H, Radvanyi F, Duros C, Saquet C, Blanche
M, Junien C, Gourdon G. 1998. Somatic instability of the CTG repeat in mice
transgenic for the myotonic dystrophy region is age dependent but not correlated
to the relative intertissue transcription levels and proliferative capacities. Hum
Mol Genet 7:1285–1291.
Lin Y, Hubert Jr L, Wilson JH. 2009. Transcription destabilizes triplet repeats. Mol
Lin Y, Leng M, Wan M, Wilson JH. 2010. Convergent transcription through a long
Lin Y, Wilson JH. 2007. Transcription-induced CAG repeat contraction in human cells
is mediated in part by transcription-coupled nucleotide excision repair. Mol Cell
Mangiarini L, Sathasivam K, Mahal A, Mott R, Seller M, Bates GP. 1997. Instability
of highly expanded CAG repeats in mice transgenic for the Huntington’s disease
mutation. Nat Genet 15:197–200.
last cell cultures from Huntington’s disease transgenic mice. Brain Res 835:74–79.
HUMAN MUTATION, Vol. 34, No. 1, 157–166, 2013
Manley K, Shirley TL, Flaherty L, Messer A. 1999b. Msh2 deficiency prevents in vivo
somatic instability of the CAG repeat in Huntington disease transgenic mice. Nat
Marozzi A, Vegetti W, Manfredini E, Tibiletti MG, Testa G, Crosignani PG, Ginelli E,
Meneveri R, Dalpra L. 2000. Association between idiopathic premature ovarian
failure and fragile X premutation. Hum Reprod 15:197–202.
deficiency associated with overexpression of the MSH3 gene. Proc Natl Acad Sci
McIvor EI, Polak U, Napierala M. 2010. New insights into repeat instability: role of
RNA∗DNA hybrids. RNA Biol 7:551–558.
Mirkin SM. 2007. Expandable DNA repeats and human disease. Nature 447:932–940.
Mochmann LH, Wells RD. 2004. Transcription influences the types of deletion and
Nucleic Acids Res 32:4469–4479.
in women with premature ovarian failure. J Med Genet 35:637–640.
Nagy A, Gertsenstein M, Vintersten K, Behringer R. 2003. Manipulating the mouse
embryo: a laboratory manual. CSHL Press, Long Island, NY.
Nakamori M, Pearson CE, Thornton CA. 2011. Bidirectional transcription stimulates
expansion and contraction of expanded (CTG)∗(CAG) repeats. Hum Mol Genet
Nolin SL, Ding XH, Houck GE, Brown WT, Dobkin C. 2008. Fragile X full mutation
Genet A 146A:60–65.
J. 1991. Instability of a 550-base pair DNA segment and abnormal methylation in
fragile X syndrome. Science 252:1097–1102.
Salat U, Bardoni B, Wohrle D, Steinbach P. 2000. Increase of FMRP expression, raised
fragile X repeat expansions: a clue to the sex bias in the transmission of full
mutations? J Med Genet 37:842–850.
Schumacher S, Pinet I, Bichara M. 2001. Modulation of transcription reveals a new
mechanism of triplet repeat instability in Escherichia coli. J Mol Biol 307:39–49.
Seriola A, Spits C, Simard JP, Hilven P, Haentjens P, Pearson CE, Sermon K. 2011.
Huntington’s and myotonic dystrophy hESCs: down-regulated trinucleotide re-
peat instability and mismatch repair machinery expression upon differentiation.
Hum Mol Genet 20:176–185.
Tanaka F, Reeves MF, Ito Y, Matsumoto M, Li M, Miwa S, Inukai A, Yamamoto M,
Doyu M, Yoshida M, et al. 1999. Tissue-specific somatic mosaicism in spinal
and bulbar muscular atrophy is dependent on CAG-repeat length and androgen
receptor–gene expression level. Am J Hum Genet 65:966–973.
RJ, Hagerman PJ. 2007. CGG repeat length correlates with age of onset of motor
signs of the fragile X-associated tremor/ataxia syndrome (FXTAS). Am J Med
Genet B Neuropsychiatr Genet 144B:566–569.
RJ. 1999. Tissue heterogeneity of the FMR1 mutation in a high-functioning male
with fragile X syndrome. Am J Med Genet 84: 233–239.
Telenius H, Kremer B, Goldberg YP, Theilmann J, Andrew SE, Zeisler J, Adam S,
Greenberg C, Ives EJ, Clarke LA, et al. 1994. Somatic and gonadal mosaicism of
Tome S, Holt I, Edelmann W, Morris GE, Munnich A, Pearson CE, Gourdon G. 2009.
mice. PLoS Genet 5:e1000482.
Ueno S, Kondoh K, Kotani Y, Komure O, Kuno S, Kawai J, Hazama F, Sano A. 1995.
Somatic mosaicism of CAG repeat in dentatorubral-pallidoluysian atrophy (DR-
PLA). Hum Mol Genet 4:663–666.
Uzielli ML, Guarducci S, Lapi E, Cecconi A, Ricci U, Ricotti G, Biondi C, Scarselli B,
Vieri F, Scarnato P, et al. 1999. Premature ovarian failure (POF) and fragile X
premutation females: from POF to to fragile X carrier identification, from fragile
X carrier diagnosis to POF association data. Am J Med Genet 84:300–303.
van den Broek WJ, Nelen MR, Wansink DG, Coerwinkel MM, te Riele H, Groenen
PJ, Wieringa B. 2002. Somatic expansion behaviour of the (CTG)n repeat in
myotonic dystrophy knock-in mice is differentially affected by Msh3 and Msh6
mismatch-repair proteins. Hum Mol Genet 11:191–198.
Vianna-Morgante AM, Costa SS, Pares AS, Verreschi IT. 1996. FRAXA premutation
associated with premature ovarian failure. Am J Med Genet 64:373–375.
Wang C, Jurk D, Maddick M, Nelson G, Martin-Ruiz C, von Zglinicki T. 2009. DNA
damage response and cellular senescence in tissues of aging mice. Aging Cell
repair gene Msh2 modifies the timing of early disease in Hdh(Q111) striatum.
Hum Mol Genet 12:273–281.
Wu Z, Falciatori I, Molyneux LA, Richardson TE, Chapman KM, Hamra FK. 2009.
Spermatogonial culture medium: an effective and efficient nutrient mixture for
culturing rat spermatogonial stem cells. Biol Reprod 81:77–86.