Down syndrome: A study of chromosomal mosaicism

Abstract

Recent data suggest that chromosome mosaicism is a possible mechanism for intrauterine and postnatal survival in cases of trisomy 18 and Turner syndrome (45X). The aim of this study was to evaluate if chromosomal mosaicism is a possible mechanism of survival in Down syndrome (DS) (trisomy 21) individuals. Mosaicism was studied by interphase fluorescence in-situ hybridization (FISH), using a specific probe for chromosome 21 (21q22.13–21q22.2) in 78 cases suspected of DS. To rule out tissue specific mosaicism, buccal cells or amniocytes were analysed in addition to blood in 20 DS cases. Thirty-three per cent of the cases studied by FISH in only peripheral blood were mosaics. In 20 cases of trisomy 21, two tissues were studied and mosaicism was not detected in either of the two tissues in 15 cases. The remaining five cases were mosaics in both the tissues analysed. Clinical comparisons in 17 DS mosaics showed a direct relationship between the percentage of trisomic cells and the degree of phenotypic manifestations. These results suggest that mechanism(s) other than mosaicism may exist for the intrauterine and postnatal survival of DS cases.

Full text

Article
499
Introduction
At least 25% of preimplantation conceptuses and 15% of
clinically recognized pregnancies are lost spontaneously. One
major reason for this high rate of pregnancy loss is
chromosome aneuploidy. A significant proportion of human
preimplantation embryos and first trimester fetuses that abort
spontaneously are chromosomally defective (Jacobs and
Hassold, 1987; Márquez et al., 2000; Munné et al., 2002).
Approximately 50% of first trimester spontaneous abortions
are trisomic, 20% are monosomic and 25% are polyploid
(Jacobs and Hassold, 1987).
Trisomy for almost every chromosome has been described in
spontaneously aborted fetuses; some trisomies are very
frequent (e.g. trisomy 15, 16, 21, 18), whereas others are rare
(e.g. trisomy 1, 5, 19). Irrespective of their frequency of
occurrence, one common feature of all trisomies is that they
act as lethal mutations to the developing embryo or fetus,
although some trisomic fetuses do survive to term. It has been
estimated that of conceptuses with trisomy 13 or 18, less than
5% survive till term, whereas about 20–25% of trisomy 21
conceptuses are live born (Kalousek et al., 1989).
What makes these few trisomic conceptions different from
others in terms of their intrauterine survival? Current dogma
holds that under the pressure of natural selection, most
aneuploid conceptions are aborted spontaneously, but natural
selection does not prevail when mosaicism is operative; thus
most live born chromosomally aneuploid fetuses may be
mosaics (Hook and Warburton, 1983; Kalousek et al., 1989;
Modi et al., 1999). Using interphase fluorescence in-situ
hybridization (FISH), it has been shown that all live born cases
of trisomy 18 have a coexisting normal diploid (disomy 18)
cell line, indicating that mosaicism is a possible mechanism of
intrauterine and postnatal survival of infants with trisomy 18
(Modi et al., 1999). However, in the same study, only 20% of
Down syndrome (DS) patients studied were mosaics. This
frequency of mosaicism was lower than that observed in cases
of trisomy 18 (100%) and Turner syndrome (75%) (Modi et
Down syndrome: a study of chromosomal
mosaicismDeepak Modi obtained his Bachelors degree in Zoology in 1993 and obtained his Masters
degree in Animal Physiology in 1995. He completed his PhD degree in Applied Biology in
2002, the subject of which was the understanding of the molecular basis of ovarian and
testicular differentiation in human fetuses. His present research interests are in deciphering
the molecular cascade of endometrial implantation and endometrial–embryo cross talk in
primates, and the characterization of progesterone receptors on spermatozoa.
Deepak Modi1, Prajakta Berde, Deepa Bhartiya
Cell Biology Department, Research Society, Bai Jerbai Wadia Hospital for Children, Acharya Donde Marg, Parel,
Mumbai, India
1Correspondence: Primate Biology Division, National Institute for Research in Reproductive Health (NIRRH), JM
Street, Parel, Mumbai 400 012, India. Tel: +91 22 4121111; e-mail: deepaknmodi@hotmail.com
Dr Deepak Modi
Abstract
Recent data suggest that chromosome mosaicism is a possible mechanism for intrauterine and postnatal survival in cases of
trisomy 18 and Turner syndrome (45X). The aim of this study was to evaluate if chromosomal mosaicism is a possible
mechanism of survival in Down syndrome (DS) (trisomy 21) individuals. Mosaicism was studied by interphase fluorescence
in-situ hybridization (FISH), using a specific probe for chromosome 21 (21q22.13–21q22.2) in 78 cases suspected of DS.
To rule out tissue specific mosaicism, buccal cells or amniocytes were analysed in addition to blood in 20 DS cases. Thirty-
three per cent of the cases studied by FISH in only peripheral blood were mosaics. In 20 cases of trisomy 21, two tissues
were studied and mosaicism was not detected in either of the two tissues in 15 cases. The remaining five cases were mosaics
in both the tissues analysed. Clinical comparisons in 17 DS mosaics showed a direct relationship between the percentage of
trisomic cells and the degree of phenotypic manifestations. These results suggest that mechanism(s) other than mosaicism
may exist for the intrauterine and postnatal survival of DS cases.
Keywords: chromosome aneuploidy, Down syndrome, fluorescence in-situ hybridization (FISH), mosaicism, trisomy 21
R
BM
Online - Vol 6. No 4. 499–503 Reproductive BioMedicine Online; www.rbmonline.com/Article/787 on web 17 February 2003

500
Articles - Mosaicism in Down syndrome - D Modi et al.
al., 1999). It was concluded that chromosome mosaicism
might not be a possible mechanism for survival of DS infants.
However, in that study, only peripheral blood cells were
analysed for detection of mosaicism. It is likely that the
absence of mosaicism in blood does not imply its absence in
other tissues, as FISH analysis in multiple tissues of Turner
syndrome fetuses revealed mosaicism in some tissues that was
not detected in amniotic fluid cells (Ameil et al., 1996).
Similarly, it is possible that live born cases of DS may be
mosaics in tissues other than blood.
Another possible reason for the low incidence of trisomy 21
mosaics observed in the earlier study (Modiet al., 1999) could
be the failure to detect mosaicism because of technical
difficulties. In the previous study, an alpha satellite probe was
used to detect trisomy 21; through sequence homology, the
probe also cross hybridizes to the centromere of chromosome
13, yielding four hybridization signals in a normal case. This
probe has limited sensitivity in detecting low level mosaicism
as compared with other chromosome specific probes (Modi et
al., 1999). Hence it is likely that mosaicism, particularly of
low degree, may have been missed.
Therefore the purpose of the study was to analyse the ploidy
level of chromosome 21 in clinically suspected cases of DS
using a probe specific to chromosome 21. To rule out tissue
specific mosaicism cells from buccal mucosa or amniocytes
were analysed in addition to peripheral blood. The aim of the
study was to detect trisomy 21 mosaicism more sensitively and
hence comment on whether or not mosaicism is a possible
mechanism of intrauterine survival of trisomy 21 infants.
Materials and methods
Seventy-eight cases with a clinical suspicion of DS were
included in this study. The inclusion criteria were those
described previously (Modi et al., 1999). Mononuclear cells
from peripheral blood or cord blood were isolated and fixed
according to standard cytogenetic protocol. Oral mucosa cells
(buccal cells) were washed twice in normal saline and fixed in
3:1 methanol:acetic acid. Aliquots of 15 ml of amniotic fluid
were collected transabdominally (16–24 weeks of gestation)
and spun at 500 gfor 10 min. The pellet (amniocytes) was
washed once in normal saline and processed according to
routine cytogenetic protocol. The specific probe for
chromosome 21 was purchased from Vysis (Richmond, UK),
and encompassed D21S259. D21S341 and D21S342 loci on
the long arm of chromosome 21 (21q22.13–21q22.2). FISH
was performed according to manufacturer’s instructions.
Briefly, the cells were spread on a silane-treated slide and air-
dried. Amniocytes and buccal cells were pretreated with
0.005% pepsin and post-fixed in 4% formaldehyde. After
incubation in 2×SSC at 37°C, the slides were dehydrated in
ethanol grades. The diluted probe was then applied on the
slides, sealed and co-denatured with the target cells at 75°C for
5 min. Hybridization was performed overnight at 37°C. The
slides were washed in 0.4×SSC at 70°C for 2 min and in 2×
SSC at room temperature for 1 min. The cells were mounted in
aquamount containing an antifade and DAPI (4′,6-diamido-2-
phenylindole) as a counterstain. At least 500 mononuclear
cells, 200 buccal cells or 100 amniocytes were scored for the
number of signals under a fluorescence microscope (Olympus
BX 60) using appropriate filter sets.
Mononuclear cells, buccal cells and amniotic fluid cells from
25 normal cases were analysed to detect hybridization
efficiency and frequency of signal distribution.
Results
The hybridization efficiency of the probe used was 99% for
mononuclear and buccal cells and 96% for amniocytes. The
lower limit for detecting a separate cell line was 2% in
mononuclear and buccal cells and 4% for amniocytes. These
values corresponded to the mean +2 SD of the false positive as
reported previously (Ameil et al., 1996; Modi et al., 1999). As
the true false negative values cannot be estimated, the upper
limit for detecting mosaicism was arbitrarily assumed as 95%.
This implied that any case showing <2% trisomic cells in
blood/buccal cells and <4% in amniocytes was considered as
normal and a non-mosaic trisomy was defined when >95%
cells were trisomic.
Figure 1 shows FISH results of a case with trisomy 21
mosaicism detected in the peripheral blood. Seventy of 78
cases with clinical suspicion of DS were found to be trisomic
Figure 1: Fluorescence in-situ
hybridization (FISH) on blood
mononuclear cells using a locus
specific 21 chromosome probe.
Note the presence of two and
three signals (arrow) indicating
mosaicism.

501
Articles - Mosaicism in Down syndrome - D Modi et al.
in peripheral blood, of which 33% (n= 23) were mosaics
(Table 1). Blood and buccal cells or amniocytes were
examined in only 20 cases of DS (Table 2). Of these, 5/20
cases were mosaics in both the tissues studied. Mosaicism was
not detected in any of the two tissues in the remaining 15
cases. The extent of mosaicism (percent trisomic cells) varied
between the two tissues in cases 1, 2 and 3. However, in the
remaining cases (two mosaic and 15 non-mosaics), the
numbers of trisomic cells were not significantly different in the
two cell types studied.
Table 3 gives the phenotypic characteristics of 17 mosaic DS
cases. There is an apparent correspondence between the
number of trisomic cells and the phenotypic manifestations
(Table 4); the number of phenotypes appears to reduce with
the fall in the number of trisomic cells (r2= 0.602; statistical
test not applied because of low numbers). The small size of the
sample, together with the subjective nature of scoring DS
phenotypes quantitatively, precluded a more detailed statistical
analysis.
Table 1. Details of live born Down syndrome cases selected in
the present study.
No. of cases selected 78
No. of cases affected 70
No. of mosaics (%) 23 (33)
Age range 2 days to 6 years
Sex ratio (male:female) 4:1
Table 2. Comparison of the percentage of trisomic cells
detected by FISH in mononuclear cells from the blood and
buccal cells or amniocytes in trisomy 21 patients.
Serial no. Blood Buccal cells or amniocytes
15036
21528
31810
43525
55045
6 99 100
7a100 98
8 100 100
9a100 96
10 99 100
11 100 100
12 100 98
13a99 95
14 97 98
15 100 95
16 100 100
17 97 98
18 99 95
19 99 96
20 100 97
aAmniocytes.
Table 3. Phenotypic characteristics of 17 mosaic DS cases.
Percentage trisomy 90 80 70 60 50 25 18 10
Patient no. 12121212123123112
Hypotonia +++++++++++–+––––
Clinodactly + + – +++++–++–++–++
Wide gap between + + – + – + ––––+–+––++
1st and 2nd toes
Broad short hands +++++?++–+++++–––
Simian crease ++++–+–++?+––+–––
Protruding tongue ++++++–+–++–+––––
Fissured tongue – + – – + + –––––––––––
Low set/small ears ++++++–+–+++–++––
Depressed ++++–+++++++++–++
nasal bridge
Small nose + – + – +++––++––+–––
Epicanthic folds ++++++–+–++–? +–+–
Flat facies +++++++–++–+++–––
Upward slant + + – – +++–++––––++–
of eyes
Accessory ? + – + – + –––?+––––––
3rd fontanelle
Brachycephaly – – ? +++–––+––+––––
Microcephaly + + ––––++–+––+++–+
Cardiac anomaly – + –––+–++?+–+–––+
+ = present; – = absent; ? = unknown/not recorded.

502
Articles - Mosaicism in Down syndrome - D Modi et al.
Discussion
In all, 33% of DS patients included in the present study
showed mosaicism using a FISH probe specific for
chromosome 21. This number is higher than that reported
earlier (20%) (Modi et al., 1999) using a common alpha
satellite 13/21 probe (20%). It is interesting to note that the
frequency of mosaicism as detected by FISH is comparatively
higher in cases of Turner syndrome (75%) and trisomy 18
(100%) than that observed for DS in the present study
(Fernandez et al., 1996; Modi et al., 1999). Thus, it appears
that chromosomal mosaicism is possibly not a common
phenomenon in cases of Down syndrome.
However, absence of mosaicism in blood does not exclude
mosaicism in other tissues. Tissue specific mosaicism has been
reported in cases of trisomy 18, trisomy 13 and Turner
syndrome (Kalousek et al., 1989; Ameil et al., 1996; D Modi,
unpublished data). To verify this possibility, buccal cells or
amniotic fluid were analysed by FISH in 20 cases of DS and
the results were compared with those obtained from blood. As
evident from Table 2, the degree of mosaicism varied between
the two tissues in some mosaic individuals, but none of the
non-mosaic cases (in blood) was found to be mosaic in the
alternate tissue investigated. Mosaicism was not detected by
FISH in the cells obtained from urine of four non-mosaic cases
of DS (D Modi, unpublished data). These findings corroborate
the results of Kalousek et al. (1989), who did not detect
mosaicism in any of the embryonic or extra-embryonic tissues
of trisomy 21 newborns. Thus, it appears that mosaicism is
probably not a common event in cases of trisomy 21, and it is
tempting to speculate that a mechanism other than mosaicism
exists to facilitate the survival of these fetuses.
Chromosome 21 is one of the smallest chromosomes in the
human genome. Recent sequencing data have shown that
chromosome 21 contains approximately 225 genes in the 33.8
Mb region sequenced (Hattori et al., 2000). However, in
comparison, in chromosome 22, which is the same size as
chromosome 21, the 33.4 Mb region sequenced has 545 genes
(Dunham et al., 1999). These results imply that chromosome
21 is gene poor as compared with chromosome 22. Thus, it is
possible that due to the low density of genes on chromosome
21, trisomy 21 is one of the most common viable non-mosaic
autosomal trisomies.
The survival of an embryo largely depends on a favourable
intrauterine anatomical, biochemical and physiological milieu.
It has been speculated that the intrauterine milieu buffers the
environmental insults and stochastic errors that the embryo
faces during development, thereby preventing the
development of serious birth defects (Shapiro, 1989, 1994).
Furthermore, it has been suggested that the loss of genetic
balance (e.g. a trisomy) predisposes the embryo to the
traumatic factors that lead to the formation of congenital
malformations and probably even embryo death (Shapiro,
1989, 1994). It has been previously shown that maternal
factors such as race, fever, alcohol use and exposure to
cigarette smoke during pregnancy influences the phenotypic
manifestations in DS (Khoury and Erickson, 1992). Thus,
developmental instability and altered homeostasis of the
aneuploid embryos increase its susceptibility to the
environmental factors, resulting in serious malformations and
even death. In this context, it is possible that DS fetuses
(although aneuploid) are less susceptible to the environmental
insults and stochastic errors faced in utero as compared with
other aneuploid individuals (e.g. trisomy 18, monosomy X),
resulting in the intrauterine and postnatal survival of these
infants even in absence of mosaicism. However, this
hypothesis needs to be exhaustively tested.
In accordance with previous results in cases of trisomy 18
(Modi et al., 1999), in the present study too, a relationship was
found between the number of trisomic cells in blood and the
phenotypic manifestations in DS cases. Although the numbers
of cases studied are too limited for rigorous statistical analysis,
a positive correlation (r2= 0.602) between the number of
trisomic cells and the phenotypic expression was observed.
Maximum phenotypes were noted when the number of
trisomic cells in blood was >50%, whereas the number of
phenotypes was minimal in mosaic individuals where the
percentage of trisomic cells was <20%. These results further
confirm the long-existing notion that mosaic aneuploid
individuals are clinically advantaged over non-mosaics
(Benda, 1969; Percy et al., 1993). However, detailed clinical
assessment and long-term follow-up of these cases will be
required for better understanding of this relationship and
making use of these findings for counselling the affected
patients especially mosaics.
The results here also recommend the use of sensitive
molecular techniques such as interphase FISH in clinical
practice for diagnosis of chromosomal disorders, as FISH
permits identification of the aneuploidy even when it is present
in low grade (low level mosaicism) which may not be detected
by using conventional karyotyping. This has particular
relevance in terms of clinical management of the patients and
counselling of the parents, especially for prenatal diagnosis in
the next pregnancy.
Table 4. Numerical relationship between percentage of
trisomic cellsaand number of phenotypes in mosaic DS casesa.
Trisomic cells (%) Number of phenotypes present
90 13
90 15
80 9
80 12
70 11
70 15
60 8
60 10
50 6
50 12
50 11
25 4
25 10
25 9
18 3
10 5
10 5
aData derived from Table 3.

503
Articles - Mosaicism in Down syndrome - D Modi et al.
In conclusion, the frequency of mosaicism as studied by FISH
in DS individuals is greater (33%) than that reported
previously (20%). However, since this frequency is still lower
than that reported for trisomy 18 (100%) and Turner syndrome
cases (75%), it is likely that different fetoprotective
pathway(s) exist which facilitate the survival of DS
individuals.
Acknowledgements
This work was supported by a grant from the Department of
Biotechnology, Ministry of Health and Sciences, India. We
thank Dr Sudha G Gangal (Director) for her valuable
suggestions. We are very grateful to the clinical staff of our
Genetic clinic.
References
Ameil A, Kidron D, Kedar I et al. 1996 Are all phenotypically
normal Turner fetuses mosaics? Prenatal Diagnosis 19, 791–795.
Benda CE 1969 Down’s syndrome. Mongolism and its Management.
Grune and Stratton, New York, pp. 121–133.
Dunham I, Collins JE, Bruskiewich R et al. 1999 The DNA sequence
of chromosome 22. Nature 402, 489–495.
Fernandez R, Mendez J, Pasaro E 1996 Turner syndrome: a study of
chromosomal mosaicism. Human Genetics 98, 29–35.
Hattori M, Fujiyama A, Taylor TD et al. 2000 The DNA sequence of
human chromosome 21. Nature 405, 311–319.
Hook EB, Warburton D 1983 The distribution of chromosomal
genotypes associated with Turner’s syndrome: live birth
prevalence and evidence for dismissed fetal mortality and
severity in genotypes associated with structural abnormalities or
mosaicism. Human Genetics 64, 24–27.
Jacobs P, Hassold T 1987 Chromosome abnormalities: etiology in
abortions and livebirths. In: Vogel F, Sperling K (eds) Human
Genetics. Springer, Berlin and Heidelberg, pp. 233–245.
Kalousek DK, Barrett IJ, McGillivray BC 1989 Placental mosaicism
and intrauterine survival of trisomies 13 and 18. American
Journal of Human Genetics 44, 338–343.
Khoury MJ, Erickson LD 1992 Can maternal risk factors influence
the presence of major birth defects in infants with Down
syndrome? American Journal of Medical Genetics 43,
1016–1022.
Márquez C, Sandalinas M, Bahce M et al. 2000 Chromosome
abnormalities in 1255 cleavage-stage human embryos.
Reproductive BioMedicine Online 1, 17–27.
Modi D, Kher A, Parikh A, Bhartiya D 1999 Chromosomal
mosaicism: evaluation by fluorescence in situ hybridization.
Medical Science Research 27, 813–815.
Munné S, Sandalinas M, Escudero T et al. 2002 Chromosome
mosaicism in cleavage-stage human embryos: evidence for a
maternal age effect. Reproductive BioMedicine Online 4,
223–232.
Percy ME, Markovic VD, Dalton AJ et al. 1993 Age associated
chromosome 21 loss in Down syndrome: possible relevance to
mosaicism and Alzheimer disease. American Journal of Medical
Genetics 45, 584–588.
Shapiro BL 1989 The pathogenesis of aneuploid phenotypes: the
fallacy of explanatory reductionism. American Journal of
Medical Genetics 33, 146–150.
Shapiro BL 1994 The environmental basis of the Down syndrome
phenotype. Developmental Medicine and Child Neurology 36,
84–90.
Received 14 October 2002; refereed 1 November 2002; accepted 13
January 2003.