Trisomy 21 Mosaicism: We May All Have a Touch of Down Syndrome

Abstract

Ever increasing sophistication in the application of new analytical technology has revealed that our genomes are much more fluid than was contemplated only a few years ago. More specifically, this concerns interindividual variation in copy number (CNV) of structural chromosome aberrations, i.e. microdeletions and microduplications. It is important to recognize that in this context, we still lack basic knowledge on the impact of the CNV in normal cells from individual tissues, including that of whole chromosomes (aneuploidy). Here, we highlight this challenge by the example of the very first chromosome aberration identified in the human genome, i.e. an extra chromosome 21 (trisomy 21, T21), which is causative of Down syndrome (DS). We consider it likely that most, if not all, of us are T21 mosaics, i.e. everyone carries some cells with an extra chromosome 21, in some tissues. In other words, we may all have a touch of DS. We further propose that the occurrence of such tissue-specific T21 mosaicism may have important ramifications for the understanding of the pathogenesis, prognosis and treatment of medical problems shared between people with DS and those in the general non-DS population.

Full text

Fax +41 61 306 12 34
E-Mail karger@karger.ch
www.karger.com
Cytogenet Genome Res
DOI: 10.1159/000346028
Trisomy 21 Mosaicism: We May All Have a
Touch of Down Syndrome
M.A. Hultén a, c J. Jonasson e E. Iwarsson c, d P. Uppal b, c S.G. Vorsanova f–h
Y.B. Yurov f–h I.Y. Iourov f, g
a Warwick Medical School, University of Warwick, Warwick , and
b Imperial College School of Medicine, London , UK;
Departments of
c Molecular Medicine and Surgery, Karolinska Institutet and
d Clinical Genetics, Karolinska
University Hospital, Stockholm , and
e Department of Clinical and Experimental Medicine, Linköping University,
Linköping , Sweden; f Institute of Pediatrics and Children Surgery, Rosmedtechnologii,
g National Research Center of
Mental Health, Russian Academy of Medical Sciences, and
h Moscow City University of Psychology and Education,
Moscow , Russia
ing of the pathogenesis, prognosis and treatment of medical
problems shared between people with DS and those in the
general non-DS population. Copyr ight © 2013 S. Karger AG, Ba sel
Recently there has been considerable focus on the ap-
plication of new technology in the identification of sub-
microscopic structural chromosome aberrations, such as
microdeletions and microduplications [Conrad et al.,
2010; Stankiewicz and Lupski, 2010; Lichtenbelt et al.,
2011; Mills et al., 2011; Poot, 2011; Forsberg et al., 2012].
Most of this work has been performed on DNA samples
extracted from blood samples, under the assumption that
the majority, if not all, tissues of the affected subjects
would carry the same abnormality. By comparison, little
attention has been given to the possible occurrence of tis-
sue-specific mosaicism, which could have significant
clinical implications [Dumanski and Piotrowski, 2012].
This lack of knowledge does in fact not only concern
structural, but also numerical chromosome aberrations
[Jackson-Cook, 2011]. We review this situation, using the
example of the very first chromosome aberration identi-
fied in the human genome, i.e. an extra chromosome 21
Key Words
Chromosome aberration ⴢ Down syndrome ⴢ Fetus ⴢ
Gonad ⴢ Mosaicism ⴢ Trisomy 21
Abstract
Ever increasing sophistication in the application of new ana-
lytical technology has revealed that our genomes are much
more fluid than was contemplated only a few years ago.
More specifically, this concerns interindividual variation in
copy number (CNV) of structural chromosome aberrations,
i.e. microdeletions and microduplications. It is important to
recognize that in this context, we still lack basic knowledge
on the impact of the CNV in normal cells from individual tis-
sues, including that of whole chromosomes (aneuploidy).
Here, we highlight this challenge by the example of the very
first chromosome aberration identified in the human ge-
nome, i.e. an extra chromosome 21 (trisomy 21, T21), which
is ca us at ive of Do wn sy nd rom e ( DS ). W e co ns id er it l ik el y th at
most, if not all, of us are T21 mosaics, i.e. everyone carries
some cells with an extra chromosome 21, in some tissues. In
other words, we may all have a touch of DS. We further pro-
pose that the occurrence of such tissue-specific T21 mosa-
icism may have important ramifications for the understand-
Publish ed online: January 10, 2013
Prof. Maj A. Hultén
Warwick Medical School
University of Warwick
Covent ry CV4 7A L (UK)
E-Mail maj.hulten  @  ki.se
© 2013 S. Ka rger AG, Basel
1424–8581/13/0000–0000$38.00/0
Accessible online at:
www.karger.com/cgr

Hultén  /Jonasson  /Iwarsson  /Uppal  /
Vor s an ov a
 /Yurov  /Iourov 
Cytogenet Genome Res
2
(trisomy 21, T21) causative of Down syndrome (DS)
[Lejeune et al., 1959]. In this setting, it is particularly im-
portant to note that even the most sophisticated microar-
ray technology will not allow detection of copy number
variations (CNV) mosaicism occurring in fewer than 5%
of cells. Currently, the only technology available for the
reliable identif ication of lower grades of mosa icism i s flu-
orescence in situ hybridization (FISH). Uniquely, this
technology allows the identification of CNV in individu-
al cell nuclei without DNA extraction. Here the limitation
in identifying low-grade mosaicism is the number of cells
to be analyzed by fluorescence microscopy and, thus, the
labour involved in this [see e.g. Hultén et al., 2010a; Vor-
sanova et al., 2010].
High-Grade T21 Mosaicism in DS Cases
It is now well recognized that DS is the most common
genetic reason for learning disability, estimated to affect
around 1/550–1/1,000 newborns worldwide [Oster-Gra-
nite et al., 2011]. Numerous studies have implicated that
the majority of people diagnosed as having the typical
features of DS carry the extra chromosome in all their
body cells. A minority (around 1–2%) are high-grade T21
mosaics, most commonly represented by an admixture
with a small proportion of cells (around 10–20%) having
the normal chromosome complement.
As might be expected, the severity of the clinical pic-
ture in such DS cases is related to the degree of T21 mo-
saicism, usually ascertained by examination of in vitro
cultured blood lymphocytes and, more rarely, uncul-
tured tissue samples, such as buccal smears [Papavassi-
liou et al., 2009; Shin et al., 2010].
Low-Grade T21 Mosaicism in Borderline DS Cases
Low-grade T21 mosaicism was first recognized by
Clarke et al. [1961, 1963], who performed a detailed chro-
mosome analysis of several tissue samples from a girl
w it h a ppar entl y n or ma l i nt el li genc e a nd on ly su bt le fa ci al
dysmorphism as seen with DS. Here, they identified T21
cells in a blood culture (14%), a bone marrow sample
(17%) as well as in 2 skin cultures (32 and 38%).
Since then a large number of similar individual case
reports have been published. The indication for the ex-
tended chromosome analysis has varied. The incidence of
T21 cells is usually investigated by analysis of in vitro cul-
tured blood lymphocytes and is in the order of a couple
of hundred cells. This is exemplified by a recent report on
a patient with young-onset dementia [Ringman et al.,
2008]. Interestingly, however, another recent case of a
new-born girl with only subtle features of DS showed an
absence of T21 cells in a traditional blood lymphocyte
culture (0/22 metaphases) but 31% T21 interphase cell nu-
clei in buccal smears by FISH. Multiple duplications
along ch romosome 21 were re vea led by microarray geno -
typing of blood DNA [Leon et al., 2010].
Low-Grade T21 Mosaicism in the General Population
Our own interest has focused specifically on the oc-
currence of even lower-grade T21 mosaicism in the gen-
eral non-DS population, where there has been no specific
reason to suspect T21 mosaicism [Hultén et al., 2008,
2010a, b; review in Iourov et al., 2008, 2010]. The inci-
dence of such low-grade T21 mosaicism, i.e. occurring in
less than 0.5–1.0% of cells, can currently only be verified
by fluorescence microscopy analysis of interphase nuclei
in the different tissue samples, labelled with chromo-
some-specific DNA probes ( fig.1 ) [review in Vorsanova
et al., 2010]. The number of cells to be analy zed must also
be high, i.e. in the order of thousands rather than hun-
dreds. Due to these technological difficulties, it comes as
no surprise that to date only a limited number of tissues
i n a sm al l n um be r o f s ubj ec ts h ave be en in ve st ig at ed , doc -
umenting the occurrence of T21 mosaicism.
One specific aspect of this concerns the identification
of parental germinal T21 mosaicism in DS. Using FISH
to record the incidence of T21 mosaicism in ovarian sam-
ples from 20 female fetuses, where termination of preg-
nancy was performed for a social reason, we have con-
cluded that most, if not all women are germinal T21 mo-
saics [Hultén et al., 2008 and unpubl. observations]. We
have further proposed that the degree of germinal T21
mosaicism, arising in the early stages of maternal oogen-
esis, may be of crucial importance in determining the
likelihood of having a child with T21 DS. The situation
in human males is more complex, as foetal testicular T21
mosaicism is very rare by comparison to foetal ovarian
T21 mosaicism [Hultén et al., 2008, 2010b]. On the other
hand, disomy 21 mosaicism in sperm is common and
generally agreed to be around 1/1,000 [table2 in Hultén
et al., 2010b; Tempest, 2011; Templado et al., 2011].
In terms of somatic mosaicism, we have paid particu-
lar attention to the increased risk for people with DS to
get Alzheimer’s disease at an early age. We have previ-
ously showed that in cases ascertained from the non-DS

T21 Mosaicism Cytogenet Genome Res
3
general population, who have been diagnosed with
Alzheimer’s Disease, the incidence of T21 cell nuclei in
brain autopsy samples is more than 10 times that in age-
and sex-matched controls [review in Iourov et al., 2008,
2010]. Using the technology described in Iourov et al.
[2012], we have now extended this type of investigation to
5,000 cases in the non-DS general population, analyzing
more than 5,000 interphase nuclei from each of 2 differ-
ent somatic tissues, i.e. chorionic villi, fetal brain and
postmortem brain samples [Iourov et al., unpubl. obser-
vations]. We then documented T21 in interphase cell nu-
cl ei r angi ng f rom 0.29 t o 0. 70%, i.e. in chor ion ic v il li (ear-
ly 0.39 and late 0.70%), fetal brain (0.30%) and autopsy
brain samples (children 0.29 and adults 0.42%).
The Wider Implications
There are wider implications concerning the occur-
rence of T21 mosaicism, for example in understanding
the prognosis of medical problems shared between DS
and non-DS populations. One of the outstanding ques-
tions here is if any such conditions could be caused (or as
the case may be hindered) by different degrees of T21 mo-
saicism in the relevant tissue samples amongst the non-
DS general population [Hultén et al., 2010a].
Trisomy mosaicism can arise either by losing one
chromosome 21, so-called ‘somatic rescue’ in a T21 zy-
gote, or by nondisjunction at subsequent cell divisions in
a normal disomy 21 zygote. It seems likely somatic ‘aneu-
ploidization’ is an on-going process throughout develop-
ment that only affects the phenotype above a certain
threshold of T21 mosaicism in various tissues [Iourov et
a l., 2 00 8]. From th is po int of v ie w, i t wou ld in fa ct be mor e
correct to say that we all have a touch of an acquired rath-
er than an inborn form of DS. Nevertheless, the pheno-
typic effect of the T21 cells in the different tissues, depen-
dent on the degree of T21 mosaicism, is likely to be the
same.
It is well established that those with DS are predis-
posed to a number of medical conditions. We have re-
cently summarized available literature concerning T21
mosaicism in relation to childhood leukaemia, solid can-
cers and Alzheimer’s disease [Hultén et al., 2010a, Bo-
rysov et al., 2011; Williams et al., 2011]. T21 mosaicism
has been seen to dramatically increase the incidence of
transient acute myeloid leukaemia, acute lymphocytic
leukaemia and testicular tumours [Hultén et al., 2010a].
Another intriguing aspect to note concerns the DS popu-
lation being predisposed to Alzheimer’s disease but, at
the same time, protected from the common medical con-
dition atherosclerosis and certain types of solid malig-
nancies, such as breast cancer [Draheim et al., 2010; Hul-
tén et al., 2010a; Borysov et al., 2011; Tabares-Seisdedos
et al., 2011]. There are, however, a large number of other
medical problems affecting the DS population [Roizen,
2010], where future research may bring new light on the
potential role of T21 mosaicism in individual tissue sam-
ples for the development and treatment of these condi-
tions.
Fig. 1. FISH images of chromosome 21 in interphase cell nuclei.
Top panel: Images of foetal ovarian cell nuclei, using 2 chromo-
some 21-specific probes located near the end of 21q, showing one
normal disomy 21 and one T21 nucleus, illustrating normal fe-
male T21 germinal mosaicism (0.54%). Middle panel: Images of
brain cel l nuclei from an autopsy of a young normal control, using
a chromosome 21-specific multicolor banding probe, showing
one normal disomy 21 and one T21 nucleus, illustrating normal
brain cell T21 mosaicism (0.29%). Bottom panel: Images of brain
cell nuclei from an autopsy of a patient with Alzheimer’s disease,
using a chromosome 21-specific multicolor banding probe, show-
ing 2 normal disomy 21 and 1 T21 nucleus, illustrating Alzhei-
mer-associated brain cell T21 mosaicism (11%). Top and middle
panels are revised from Hultén et al. [2010a].

Hultén  /Jonasson  /Iwarsson  /Uppal  /
Vor s an ov a
 /Yurov  /Iourov 
Cytogenet Genome Res
4
References
Borysov SI, Granic A, Padmanabhan J, Walczak
CE, Potter H : Alzheimer A ␤ di srupts the mi-
totic spindle and directly inhibits mitotic
microtubule motors. Cell Cycle 10:
1397–
1410 (2011).
Clarke CM, Edwards JH, Smallpeice V: 21-triso-
my/normal mosa icism in an intelligent child
with some mongoloid characters. Lancet 1:
1028–1030 (1961).
Clarke CM, Ford CE, Edwards JH, Smallpeice V:
21 trisomy/normal mosaicism in an intelli-
gent child with some mongoloid characters.
Lancet 2:
1229 (1963).
Conrad DF, Bird C, Blackburne B, Lindsay S,
Mamanova L, et al: Mutation spectrum re-
vealed by breakpoint sequencing of human
germline CNVs. Nat Genet 42:
385–391
(2010).
Dr aheim CC, G eijer JR, Dengel DR : Comparison
of intima-media thickness of the carotid ar-
tery and cardiovascular disease risk factors
in adults with versus without the Down syn-
drome. Am J Cardiol 106:
1512–1516 (2010).
Dumanski JP, Piotrowski A: Structural genetic
variation in the context of somatic mosa-
icism. Met hods Mol Biol 838:
249–272 (2012).
Forsberg L, Rasi C, Razzaghian HR, Pakalapati
G, Waite L, et al: Age-related somatic struc-
tural changes in the nuclear genome of hu-
man blood cells. Am J Hum Genet 90:
217–
228 (2012).
Hultén MA, Patel SD, Tankimanova M, West-
gren M, Papadogiannakis N, et al: On the
origin of trisomy 21 Down syndrome. Mol
Cytogenet 1:
21 (2008).
H ul tén MA , J ona ss on J, N ord gr en A, Iwa rs so n E:
Germinal and somatic trisomy 21 mosa-
icism: how common is it, what are the impli-
cations for individual carriers and how does
it come about? Curr Genomics 11:
409–419
(2010a).
Hultén MA, Patel SD, Westgren M, Papadogi-
annakis N, Jonsson AM, et al: On the pater-
nal origin of trisomy 21 Down syndrome.
Mol Cytogenet 3:
4 (2010b).
Iourov IY, Vorsanova SG, Yurov YB: Chromo-
somal mos aicism goes globa l. Mol Cytogenet
1:
26 (2008).
Iourov IY, Vorsanova SG, Yurov YB: Somat ic ge-
nome variations in health and disease. Curr
Genomics 11:
387–396 (2010).
Iourov IY, Vorsanova SG, Svetlana B, Yurov YB:
Single cell genomics of the brain: focus of
neuronal diversity and neuropsychiatric dis-
eases. Curr Genomics 13:
477–488 (2012).
Jackson-Cook C: Constitutional and acquired
autosomal a neuploidy. Clin Lab Med 31:
481–
511 (2011).
Lejeune J, Gautier M, Turpin R: Study of somat-
ic chromosomes f rom 9 mongoloid children.
C R Hebd Seances Acad Sci 248:
1721–172 2
(195 9).
Leon E, Zou YS, Mi lunsky JM: Mosaic Down
syndrome in a patient with low-level mosa-
icism detected by microarray. Am J Med
Genet A 152A:3154–3156 (2010).
L ichtenbelt KD, Knoers N V, Schurin g-Blom GH:
From karyotyping to array-CGH in prenatal
diagnosis. Cytogenet Genome Res 135:
241–
250 (2011).
Mills RE, Walter K, Stewart C, Handsaker RE ,
Chen K, et al: Mapping copy number varia-
tion by population-scale genome sequenc-
ing. Nature 470:
59–65 (2011).
Oster-Granite ML, Pa risi MA, Abbeduto L, Ber-
lin DS, Bodine C, et al: Down syndrome: na-
tional conference on patient registries, re-
search databases, and biobanks. Mol Genet
Metab 104:
13–22 (2011).
P a pa va s si l io u P, Yor k T P, Gu rs o y N , H i ll G , N ic e-
ly LV, e t al.: The phenoty pe of persons havi ng
mosaicism for trisomy 21/Down syndrome
reflects the percentage of trisomic cel ls pres-
ent in dif ferent tissues. Am J Med Genet A
149A:573–583 (2009).
Poot M: Preface. Cytogenet Genome Res 135:
171–173 (2011).
Ringman JM, Rao PN, Lu PH, Cederbaum S:
Mosaicism for trisomy 21 in a patient with
young-onset dementia: a case report and
brief literature review. Arch Neurol 65:
412–
415 (2008).
Roizen NJ: Overview of health issues among
persons with Down syndrome, in Urbano
C (ed): International Review of Research in
Ment al Re tard ation , Heal th Issues i n Dow ns
Syndrome, pp 3–34 (Academic Press, San
Diego 2010).
Shin M, Siffel C, Correa A: Survival of children
with mosaic Down syndrome. Am J Med
Genet A 152A:800–801 (2010).
Sta nkiewicz P, Lupski JR: St ructura l variation in
the human genome and its role in disease.
Annu Rev Med 61:
437–455 (2010).
Ta bar es-S eis ded os R , Du mont N, Ba udo t A, Va l-
deras JM, Climent J, et al: No paradox, no
progress: i nverse cancer comorbidit y in peo-
ple with other complex diseases. Lancet On-
col 12:
604–608 (2011).
Tempest HG: Meiotic recombination errors, the
origin of sperm aneuploidy and clinical rec-
ommendation s. Syst Biol Reprod Med 57:
93–
101 (2011).
Te mp lad o C , V id al F, E st op A : A ne upl oi dy in h u-
man spermatozoa. Cytogenet Genome Res
133:
91–99 (2011).
Vorsanova SG, Yurov YB, Iourov IY: Human in-
terphase chromosomes: a review of available
molecular cytogenetic technologies. Mol
Cytogenet 3:
1 (2010).
Williams BA, Meyn MS, Hitzler JK: Transient
leukemia in newborns without Down syn-
drome: diagnostic and management chal-
lenges. J Pediatr Hematol Oncol 33:e261–
e263 (2011).
Conclusion
On the basis of these considerations, we believe that
T21 mosaicism is a feature shared in common between
most, if not all, of the general population. Thus, it is im-
portant to recognize that the occurrence of tissue-specif-
ic T21 mosaicism may have significant ramifications for
the understanding of the pathogenesis and treatment of
common medical problems that occur with a higher in-
cidence in the DS population in comparison to the non-
DS general population.
Acknowledgements
Our work in this area has been supported by grants from the
Wellcome Trust (061202/ZOOZ) and BBSRC (BB/C003500/1) to
M.A.H., The Swedish Research Council and Stockholm County
Council to E.I., Deutsches Luft- und Raumfahrtszentrum/Bun-
desministerium für Bildung und Forschung (RUS 09/006 and
RUS 11/002) to S.G.V., Y.B.Y. and I.Y.I.