Article

Detection of=1?Mb microdeletions and microduplications in a single cell using custom oligonucleotide arrays

Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA.
Prenatal Diagnosis (Impact Factor: 3.27). 01/2012; 32(1):10-20. DOI: 10.1002/pd.2855
Source: PubMed

ABSTRACT

High resolution detection of genomic copy number abnormalities in a single cell is relevant to preimplantation genetic diagnosis and potentially to noninvasive prenatal diagnosis. Our objective is to develop a reliable array comparative genomic hybridization (CGH) platform to detect genomic imbalances as small as ~1Mb ina single cell.
We empirically optimized the conditions for oligonucleotide-based array CGH using single cells from multiple lymphoblastoid cell lines with known copy number abnormalities. To improve resolution, we designed custom arrays with high density probes covering clinically relevant genomic regions.
The detection of megabase-sized copy number variations (CNVs) in a single cell was influenced by the number of probes clustered in the interrogated region. Using our custom array, we reproducibly detected multiple chromosome abnormalities including trisomy 21, a 1.2Mb Williams syndrome deletion, and a 1.3Mb CMT1A duplication. Replicate analyses yielded consistent results.
Aneuploidy and genomic imbalances with CNVs as small as 1.2Mb in a single cell are detectable by array CGH using arrays with high-density coverage in the targeted regions. This approach has the potential to be applied for preimplantation genetic diagnosis to detect aneuploidy and common microdeletion/duplication syndromes and for noninvasive prenatal diagnosis if single fetal cells can be isolated.

Full-text

Available from: Sau Wai Cheung
ORIGINAL ARTICLE
Detection of 1 Mb microdeletions and microduplications in a
single cell using custom oligonucleotide arrays
Weimin Bi
1
, Amy Breman
1
, Chad A. Shaw
1
, Pawel Stankiewicz
1
, Tomasz Gambin
1,2
, Xinyan Lu
1
, Sau Wai Cheung
1
, Laird G. Jackson
3
,
James R. Lupski
1
, Ignatia B. Van den Veyver
1,4
and Arthur L. Beaudet
1
*
1
Department of Molecular and Human Genetics, Baylor College of Medicine, Houston, TX, USA
2
Institute of Computer Science, Warsaw University of Technology, Warsaw, Poland
3
Department of Obstetrics and Gynecology, Drexel University College of Medicine, Philadelphia, PA, USA
4
Department of Obstetrics and Gynecology, Baylor College of Medicine, Houston, TX, USA
*Correspondence to: Arthur L. Beaudet. E-mail: abeaudet@bcm.tmc.edu
ABSTRACT
Objective High resolution detection of genomic copy number abnormalities in a single cell is relevant to
preimplantation genetic diagnosis and potentially to noninvasive prenatal diagnosis. Our objective is to develop a
reliable array comparative genomic hybridization (CGH) platform to detect genomic imbalances as small as ~1 Mb in
a single cell.
Methods We empirically optimized the conditions for oligonucleotide-based array CGH using single cells from
multiple lymphoblastoid cell lines with known copy number abnormalities. To improve resolution, we designed
custom arrays with high density probes covering clinically relevant genomic regions.
Results The detection of megabase-sized copy number variations (CNVs) in a single cell was inuenced by the number
of probes clustered in the interrogated region. Using our custom array, we reproducibly detected multiple
chromosome abnormalities including trisomy 21, a 1.2 Mb Williams syndrome deletion, and a 1.3 Mb CMT1A
duplication. Replicate analyses yielded consistent results.
Conclusion Aneuploidy and genomic imbalances with CNVs as small as 1.2 Mb in a single cell are detectable by array
CGH using arrays with high-density coverage in the targeted regions. This approach has the potential to be applied for
preimplantation genetic diagnosis to detect aneuploidy and common microdeletion/duplication syndromes and for
noninvasive prenatal diagnosis if single fetal cells can be isolated. © 2012 John Wiley & Sons, Ltd.
Funding sources: None
Conicts of interest: None declared
Supporting information may be found in the online version of this article.
INTRODUCTION
Single-cell array comparative genomic hybridization (CGH)
provides information on genome copy number changes at
the single cell level, which is of great importance in both cancer
genetics and clinical diagnosis. This single cell technology
facilitates the studies of tumor heterogeneity, micrometastases,
and minimal residual disease. In addition, the application in
preimplantation diagnosis allows a comprehensive chromosome
analysis including aneuploidy analysis of all 24 chromosomes.
Furthermore, single-cell analysis may open up new opportunities
for noninvasive prenatal genetic diagnosis.
It has been well documented that fetal cells, cell-free
fetal (cff) DNA, and RNA are present in the peripheral
blood circulation of pregnant women.
1
In addition, fetal
tro phoblasts have be en ident ied in endocervical specimens
at an early stage of gestation.
2,3
Cell-free fetal DNA is
relatively abundant in the maternal circulation but it is
fragmented and not physically separated from cell-free
maternal DNA. In contrast, the genomic DNA of circulating
intact fetal cells represents a complete fetal genome that is
physically separated from maternal cellular and cell-free
DNA. Although circulating fetal cells are very rare, at about
one cell per ml of maternal blood,
4
their genetic information
is accessible by single cell technologies,
5
such as whole-
genome amplication (WGA), which can generate sufcient
DNA for array CGH.
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
DOI: 10.1002/pd.2855
Page 1
Array-based CGH (array CGH) has been used successfully in
prenatal diagnosis for high resolution genome analyses and
detection of numerical and unbalanced structural chromosome
abnormalities.
6,7
When array CGH is applied to prenatal
diagnosis, detection rates of clinically signicant abnormalities
vary from 810%.
6
Furthermore, array CGH can detect signicant
aberrations undetectable by karyotype in up to 4.3% of prenatal
studies
6,7
(our unpublished data on 750 cases). To date, denitive
prenatal diagnosis of fetal chromosomal abnormalities using
either modality karyotype or array CGH still requires the
use of invasive procedures to collect amniotic uid or
chorionic villi, which are associated with a low risk of mis-
carriage. Therefore, although the current American Congress
of Obstetricians and Gynecologists (ACOG) guidelines
recommend that Invasive diagnostic testing for aneuploidy
should be available to all women, regardless of maternal
age (ACOG practi ce bulletin 88, 2007), prenatal genetic
diagnosis is still offered primarily to women with
pregnancies at increased risk for aneuploidy because of
advanced maternal age, abnormal serum screen or
ultrasound results, or a combination of these conditions.
Prenatal genetic diagnosis would likely be more widely
implemented if genetic testing utilizes fetal materials
collected by a noninvasive procedure.
High resolution human genome analyses by array CGH
on single cells has been implemented during the past 5 years.
Spits and colleagues detected a 34 Mb deletion in single
lymphoblastoid cells and aneuploidy in single broblasts and
blastomeres using a bacterial articial chromosome (BAC)
array.
8
Fiegler et al. reported detection of an 8.3 Mb deletion
in a single cell from a tumor cell line and a 10.8 Mb deletion
in a single cell from a patient cell line on a whole-genome
tiling path BAC array.
9
The currently reported highest
resolution of array. CGH on single-cell DNA is 2.6 to 3.0 Mb,
and was obtained using tiling oligo arrays and a new
algorithm to analyze the raw array data.
10
Submicroscopic
chromosomal rearrangements can lead to a wide variety of
serious clinical manifestations and have been proven to
contribute signicantly to the cause of mental retardation.
Most microdeletion or microduplication syndromes involve
genomic rearrangements of 1 Mb to 4 Mb in size.
11
For
example, the CNVs associated with the two most frequent
microdeletion syndromes, DiGeorge syndrome (DGS; MIM #
188400) and Williams syndrome (WS; MIM# 194050), are
deletions of ~3 Mb and 1.5 Mb, respectively. Therefore, robust
methods for high-resolution genome analysis are an
important initial goal to achieve reliable detection of common
genomic aberrations (i.e. Mb size CNV), in addition to
aneuploidy, from dissected blastomeres of preimplantatio n
embryos or single cells that can be noninvasively retrieved
from the maternal blood during pregnancy. The objectives
for this study were to develop and optimize me thods to
reliably perform WGA for detection of genomic imbalances
of 1MbbyarrayCGHusingsinglecells.
MATERIALS AND METHODS
Cell lines
Lymphoblastoid cell lines established from peripheral blood
samples were used to optimize the conditions for single-cell
array CGH. One cell line was from a patient with trisomy 21
and the others each contained one genomic disorder-
associated submicroscopic copy number change. Prior to their
use for this study, array CGH using an Agilent high resolution
custom array was performed to conrm the size and genomic
location of the copy number changes on the cell lines with
aPraderWilli syndrome (PWS; MI M# 176270) deletion, a
SmithMagenis syndrome (SMS; MIM# 182290) deletion, a
DGS deletion, a WS deletion, or a CharcotMarieTooth 1A
disease (CMT1A; MIM# 118220) duplication. Copy number
changes in chromosome 17 in a cell line with a Potocki
Lupski syndrome (PTLS; MIM# 610883) duplication were
studied using a customized Agilent DNA array with a high
density of probes in chromosome 17p.
12
The case from which
the MECP2 duplication-cell line was made has been
published.
13
We also used a male and female cell line as
references for array CGH, in which no genomic imbalances
of 1 Mb in size were detected.
Isolation of single cells
Approximately 0.5 ml of cell suspension from cultured cells was
centrifuged at 1000 rpm for 10 min. The cells were washed
twice in 5 ml 1 phosphate-buffered saline (1 PBS) buffer.
Cell pellets were resuspended in 1 ml of 1 PBS and 20 to
100 μl of the cell suspension was transferred to a 35-mm Petri
dish containing 2 ml 1 PBS. One single cell was picked into
l of 1 PBS solution inside a 100-mm stripper tip using a
micropipette (MidAtlantic Diagnostics, Mount Laurel, NJ)
under an Olympus CK30 inverted phase contrast microscope
with 40 magnication by hand without a micromanipulator.
The single cell was transferred to a 35-mm Petri dish
containing 2 ml 1 PBS for further washing and conrmation
of the presence of the single cell. Then that single cell was
picked into 1 μl of 1 PBS solution again using a micropipette
and transferred to the bottom of a 0.2 ml PCR reaction tube on
ice. Each single cell was picked using a new micropipette. The
isolated cells were stored at 20
C if cell lysis was not
performed immediately thereafter.
Whole genome amplication
Single-cell WGA was performed using the PicoPlex single cell
WGA kit (Rubicon Genomics, Inc. Ann Arbor, MI) according
to the manufacturers instructions with slight modications.
The WGA procedures consist of three steps: cell lysis and DNA
fragmentation, preampli cation (Pre-Amp), and amplication.
For the rst step, 4 μl of cell extraction buffer was added to a
0.2 ml PCR reaction tube with a single cell in 1 μl of 1 PBS
solution. After adding a 5 μl mixture of extraction enzyme
dilution buffer and cell extraction enzyme, tubes were briey
High resolution single-cell array CGH
11
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 2
spinned to bring all solutions down. Cells were lysed using a
thermal cycler by incubation at 75
C for 10 min, and DNA was
fragmented by incubation at 95
C for 4 min and then cooled
down to room temperature. For the Pre-Amp step, after adding
a 5μl mixture of PicoPlex Pre-Amp buffer and PicoPlex Pre-Amp
enzyme, the samples were mixed by a brief vortex and were
briey centrifuged. The samples were incubated at 95
C for
2 min followed by 12 cycles of the following condition: 95
C,
15 s; 15
C, 50 s; 25
C, 40 s; 35
C, 30 s; 65
C, 40 s; 75
C, 40 s, and
then cooled to 4
C. For the last step, PicoPlex amplication
enzyme and PicoPlex amplication buffer were added; samples
were briey vortexed and centrifuged. DNA was amplied in a
nal volume of 75 μl for 14 cycles consisting of a denaturation
step at 95
C for 15 s, an annealing step at 65
C for 1 min, and
an extension at 75
C for 1 min. To minimize contamination,
WGA reactions were carried out inside a PCR Workstation
(AirClean Systems) and the cell lysis and the Pre-Amp steps
were carried out in a pre-PCR room.
The PCR products were puried rst using 96-well
multiscreen HTS PCR plates (Millipore, Billerica, MA)
according to the manufacturers instructions. The puried
DNA was resuspended in 75 μl of water, and then was further
cleaned using a DNA Clean Concentrator
-25 kit (Zymo
Research, Irvine, CA) according to the manufacturers
instructions. DNA was subsequently eluted into 50 μl of water
and stored at 20
C. The WGA DNA quality and concentration
were determined using a Nanodrop spectrophotometer
(Nanodrop Inc., Wilmington, DE).
Array comparative genomic hybridization
Array CGH was performed using oligonucleotide-based
custom arrays (Agilent Technologies, Santa Clara, CA) as
described previously
14
with slight modications. Because the
size of the WGA products range from 100 bp to 1 kb, it was
not necessary to perform DNA fragmentation before labeling.
One microgram of WGA DNA was labeled for each hybrid-
ization. Test DNAs were lab eled with dCTP-Cy5 whereas
reference DNA was labeled with dCTP-Cy3. Unincorporated
nucleotides were removed using a MultiSc reen-PCR96 Filter
Plate (Mill ipore, Billerica, MA) according to manu facturers
instruction with slight modications. Briey, labeling re-
actions were mixed with 50 μl 1 Tris-EDTA (1 x TE) buffer
(pH 8.0) and transferred into a well. The plate was placed on
a Multi Scr ee nHT S vac uu m manifold (Millipore, Billerica,
MA) and vacuum was applied at 20 inHg for 10 min. The
plate was removed from the manifold and the bottom of
the plate was blotted using an absorbent material to
remove any remaining droplets. The plate was vacuumed
again for 10 min and the puried DNA was resuspended in
25 μl 1 TE buffer using a plate mixer. When only a partial
plate was used, the plate was reused for unused wells that
were two wells away from used wells.
Hybridizations were carried out at 65
C for 4072 h to
enhance the binding of WGA DNA. After washing, slides were
scanned using an Agilent Microarray Scanner (PN G2565BA).
The reference DNA was WGA DNA amplied from a single cell
of either the male or female reference cell lines. For some
experiments, we pooled WGA DNA from multiple single-cell
WGA reactions using cells from the same reference cell line.
We also tested the performance of array CGH using as a
reference WGA DNA amplied from 5 ng of genomic DNA that
was extracted from blood samples of healthy individuals.
Gender-mismatched references were used unless otherwise
indicated.
The arrays used in this study are listed in Table 1 with the
number of probes in the tested regions indicated. The BCM
V6.5 Oligo array was an Agilent 44 K array containing 44,000
oligonucleotides with coverage for each G-band at the 550
Table 1 Genomic rearrangements in the cell lines and the numbers of probes covering the involved regions in the customized Agilent
arrays
Cell line Syndrome Rearrangement Region Sex Size (Mb)
Probe numbers in arrays
BCM v6.5 NIP1 NIP2 NIP3
44 K 180 K 180 K 400 K
T21 Trisomy 21 Trisomy 21q M 32.9 622 100 k 2578 6854
WS Williams Syndrome Deletion 7q11.23 M 1.2 139 576 1892
PWS PraderWilli Syndrome Deletion 15q11q13 M 6.6 248 1302 4061
CMT1A CharcotMarieTooth disease 1A Duplication 17p12 F 1.3 106 20 k 407 2765
SMS SmithMagenis Syndrome Deletion 17p11.2 M 3.8 107 961 3335
PTLS PotockiLupski Syndrome Duplication 17p11.2 M 3.8 107 961 3335
DGS DiGeorge Syndrome Deletion 22q11.2 F 2.5 120 925 3101
MECP2 MECP2 duplication Syndrome Duplication Xq28 M 0.6 381 246 1334
The sizes of the copy number changes are based on the results of array CGH analyses on unamplied DNA as described in the Materials and Methods section. The DGS cell
line has an additional copy of chromosome 12. The PWS cell line contains gains and losses of 1 Mb in six additional genomic regions.
W. Bi et al.
12
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 3
band level and higher coverage in disease-associated regions.
14
Three customized Agilent arrays were specically designed for
this study, and were named Non-Invasive Prenatal 1 through 3
(NIP1 through NIP3). The NIP1 array is a 180 K Agilent
oligonucleotide array with a high density of probes for the
CMT1A region on chromosome 17 and for ve separate regions
on chromosome 21, each covered by ~20,000 probes (with a
probe at approximately every 80 bp). For each region, 1000
oligos with the best performance scores as determined by
eArray criteria were chosen from 1M Agilent catalog probes,
and an additional 10,000 probes were selected from all
available oligos (24M Agilent probe set). Half of those probes
were the opposite strands of the same oligos. CNV and
segmental duplication regions were excluded from the
coverage. The remainder of the genome was interrogated at
low density (70 kb/probe). The NIP2 array is a 180 K array with
high-density targeted coverage with a probe at approxi-
mately every 3.5 kb for the genomic regions of 64 common
microdeletion and microduplication syndromes and 41
subtelomeric regions. Each region was interrogated by a
similar number (~1300) of oligos, which were selected from
the 1 M Agilent catalog pr obes. The remainder of the genome
was interrogat ed at a density of about one probe every 60 kb.
The NIP3 array is a 400 K array with an even higher density of
probes within the same targeted regions as on the NIP2 array.
Every genomic disorder region-of-interest was covered by
approximately 4000 oligos selected from the 24 M Agilent
probe set. Probes that are included in segmental duplications
or overlap with other oligo probes by more than 20% were
ltered out. The probe density is about one interrogating
probe every 425 bp in the targe ted region and about one
probe every 500 kb for the remainder of the genome.
Analysis of array comparative genomic hybridization data
Microarray image les were quantied using Agilent Feature
Extraction software (version 9.5). Text le outputs containing
quantitative data were imported into the Agilent Genomic
Workbench 5.0 software (version 5.0). Data were analyzed
using the aberration detection method 2 (ADM-2) statistical
algorithm at thresholds of 3.0 or 5.0 to identify genomic
intervals with copy number changes. ADM-2 uses an iterative
procedure to identify all genomic regions for which the
weighted average by log2 ratio of the measured probe signals
is different from the expected value of 0 by more than a given
threshold. To reduce false positive calls, a lter was applied
to dene the minimum log2 ratio and the minimum number
of probes in a CNV interval.
In addition to the Genomic Workbench analyses, we
performed our own data processing in the R statistical
programming environment. This analysis included investigation
of the log ratio variance and intensity properties of the
array probes across the ensemble of hybridizations we
performed. We computed sliding window mean proles for
each array as well as for the respective disease regions
represented in each hybridization f or a variety of window
sizesfrom50to200probes.Wealsocomputedmean-
centered instances of the data set wh ere the mean value
for each indi vidual probe across all hybridizations was
subtracted and those data we re subsequently subjected t o
sliding window mean proling.
RESULTS
Genomic rearrangements in the cell lines
Lymphoblastoid cell lines with either trisomy 21 or a disease-
associated submicroscopic copy number change as listed in
Materials and Methods and as specied in tables and gure
legends were used to optimize the conditions for single-cell
array CGH (Table 1). Notably, additional copy number changes
>1 Mb in size were identied in two cell lines. Analyses of the
DGS cell line also revealed an additional copy of chromosome
12, likely resulting from a tissue culture artifact. The PWS cell
line contains copy number changes in six other genomic
regions including terminal gains in 5q, 13q, and 14q of
105 Mb, 30 Mb, and 22 Mb in size respectively, and terminal
losses in 6q, 7p, and 9p of 19 Mb, 4.5 Mb, and 28 Mb in size
respectively.
Optimization of WGA and array CGH conditions
We performed 118 single-cell WGA reactions in total, and the
WGA DNA yields after purication were higher than 2.0 μg in
86% of the cases. About 9% of WGA reactions had a yield of
0.4 μg or less, which was considered to have failed. This
complete failure was likely due to the absence of cells inside
the PCR tubes or lack of nuclei in the cells isolated. The
remaining 5% of reactions had a DNA yield between 0.4 to
2 μg. The lower than average amount of DNA could be because
of apoptosis in these cells or inefcient WGA.
To determine the appropriate reference for single cell array
CGH, we tested four individually treated single cells from the
trisomy 21 cell line and three from the CMT1A cell line. For
each cell, we performed two hybridizations with the same
WGA DNA. For one, the reference was the WGA product from
5 ng genomic DNA of a healthy individual. For the other, the
reference was the WGA product of a single cell from a reference
cell line established from a different healthy individual. All
reference WGA DNAs were gender-mismatched compared to
the test WGA DNA. Data were analyzed using the Agilent
Genomic Workbench 5.0 ADM-2 algorithm with a threshold
set at 3.0. The expected changes in chromosomes 21, 17, and
X were detected in all hybridizations. There were much fewer
false positive signals in autosomes when a single cell WGA
product was used as the reference (Table S1). When array data
were analyzed using a threshold of 5.0, the expected autosomal
copy number changes in one trisomy 21 cell and in one CMT1A
cell were detected with the single cell WGA reference but not
with the 5 ng WGA reference.
High resolution single-cell array CGH
13
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 4
We also examined the use of a pooled reference control for
the hybridization by combining 10 to 15 WGA reaction
products from distinct single cell amplications, each with a
DNA yield of 2.5 ug. We compared pooled WGA reference
DNA to single-cell WGA reference DNA by hybridizing each
to the WGA product of a single WS cell and that of a single
PWS cell. The expected copy number changes were detected
with both references using a threshold of either 3.0 or 5.0, but
fewer false positive changes were observed when the pooled
reference was used as a control (Table S1).
Our experience suggests that, compared to the WGA product
from 5 ng of genomic DNA, the use of single-cell WGA
reference DNA reduced the number of false positive calls and
improved the sensitivity to detect known copy number
changes. The detection of false-positives could be further
reduced by using pooled single-cell reference WGA products.
The data presented in the following sections used either a
single cell WGA reference or pooled reference as a
hybridization control.
Detection of aneuploidy in a single cell using the BCM V6.5 oligo
array
One goal of prenatal genetic diagnosis is to screen for
aneuploidy of all chromosomes in a fetus. Because trisomy 21
is the most common clinically relevant prenatally assessed
chromosome aberration, we used a cell line from a subject
with trisomy 21 to test our ability to detect aneuploidy in a
single cell by array CGH. Gender-mismatched references were
used for internal quality control.
Array CGH was performed on two single male cells with
trisomy 21, whose DNA was individually amplied by WGA
using the BCM V6.5 array. The reference sample was gender-
mismatched WGA-DNA amplied from reference single cells.
For both single cells, the anticipated copy number gain of
chromosome 21 and the copy number differences for the X
and Y chromosomes, because of the utilization of gender
mismatched reference, were detected using the ADM-2
algorithm and a threshold of 5.0 for data analysis (data not
shown).
When we applied an additional lter to remove all copy
number changes detected by 30 probes or by an absolute
average log ratio of 0.3, no other autosomal copy number
changes were identied. Through customized data processing,
the gain in chromosome 21 and changes in sex chromosomes
are apparent on the array CGH plots (Figure 1). Therefore,
aneuploidy such as trisomy 21 is readily detected in a single
cell by array CGH using the BCM V6.5 array.
Detection of submicroscopic changes was improved by
increasing the probe density in targeted regions
To test whether microdeletions or microduplications in a
single cell are detectable using the BCM V6.5 array, we
performed array CGH with WGA DNA from two single cells
with a CMT1A duplication, and from one single cell each with
a PWS, WS, DGS, or SMS deletion. All hybridizations were
performed with gender-mismatched reference WGA DNA. For
all but one of the CMT1A cells, the expected copy number
changes on autosomes and on the sex chromosomes were
Figure 1 Array CGH of a trisomy 21 single cell on a BCM v6.5 array. The gain in chromosome 21 was detected after customized data
analysis. The loss in chromosome X and the gain in chromosome Y were observed because gender mismatched reference WGA DNA was
used. The X-axis represents the alignment of chromosomes 1 to 22 followed by X and Y. The Y -axis shows the mean centered log ratio values
subjected to sliding window smoothing of the hybridization signals. The blue curve is a smoothing spline that interpolates the sliding window
smoothing. For the chromosome 21, the smoothing spline is separately depicted in green. The smoothing splines for the X and Y chromosomes
are depicted in red and green, respectively
W. Bi et al.
14
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 5
detected using a threshold of 3.0. However, even after ltering
out the changes detected by 30 probes or an average log ratio
0.3, the false positive calls on autosomes remained high,
ranging from 29 to 108. To reduce false positive ndings, data
were analyzed using a threshold of 5.0. With these parameters,
the false positive detections were reduced to 8. In addition,
the expected submicroscopic copy number changes were
detected in the PWS, WB, and DGS cells, but not in the CMT1A
and the SMS cells.
The BCM V6.5 array contains 622 probes covering the long
arm of chromosome 21, but only 106 probes for the 1.3 Mb
CMT1A region in 17p12. We hypothesized that the detection
of submicroscopic changes would be more reliable if the
number of probes in the targeted regions is increased. To test
whether the CMT1A duplication is detectable by increasing
oligonucleotide probes in this region, individually amplied
WGA DNAs from four different single CMT1A cells were
hybridized to a custom 180 K array (NIP1) that contains
20,000 probes for the CMT1A and its anking regions. The
17p12 gain was detected in all four experiments using a
threshold of 5.0 for data analysis (data not shown). After
further customized data processing, the gain in the CMT1A
region was readily apparent (Figure 2). These data suggested
that submicroscopic changes, including gains as small as
1.3 Mb are potentially detectable in a single cell by increasing
the density of interrogating probes in the targeted regions.
Detection of rearrangements larger than 1 Mb associated with
genomic disorders by single cell array CGH
To improve the detection of additional disease-associated
genomic imbalances in single cells, we designed a second
180 K Agilent NIP2 customized array with high-density probe
coverage for 64 known genomic disorder regions and 41
subtelomeric regions. We evaluated the ability of this array to
detect the CMT1A and PTLS duplications and the DGS, WS,
PWS, and SMS deletions. In total, we performed 14
hybridizations with WGA DNA from 12 individually amplied
single cells. Control samples were either single cell WGA
reference DNA or pooled WGA reference DNA (Table 2).
For the 14 hybridizations on the NIP2 array, all of the
expected submicroscopic changes were detected using the
ADM-2 algorithm and a threshold of 3.0. We also detected all
the expected aneuploidies and sex chromosome-imbalance
Figure 2 Array CGH of a single cell with a 1.3 Mb CMT1A duplication on the customized NIP2 array that has 20,000 probes in this region.
The detection of the 1.3 Mb copy number gain in a single CMT1A cell was clearly visualized after customized data analysis. The X-axis
represents alignment of chromosomes 1 to 22; X and Y are not shown on this rendering. The Y-axis shows the normalized log ratios of the
hybridization signals. Note that chromosome 21 takes up a disproportionately large area on the graph, because the array has ~100 K probes
covering this chromosome
Table 2 Summary of the detections of the expected genomic rearrangements by array CGH on single cells using the NIP2 180 K array
Cell lines Regions Gain or loss Size Single cells Hybridi-zations Detection rate
PWS 15q11q13 Loss 6.6 Mb 1 2 2/2
SMS 17p11.2 Loss 3.8 Mb 1 1 1/1
PTLS 17p11.2 Gain 3.8 Mb 1 1 1/1
DGS 22q11.2 Loss 2.5 Mb 3 3 3/3
CMT1A 17p12 Gain 1.3 Mb 4 4 3/4
High resolution single-cell array CGH
15
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 6
from gender-mismatched hybridizations. Under these
conditions, the number of false positive calls was high. This
problem was reduced by increasing the threshold, but this also
lowered the sensitivity. We found that the best sensitivity with
the least false positive calls was reached using the ADM-2
algorithm with a threshold set at 5.0. Using these parameters,
the expected submicroscopic copy number changes were
detected in almost all the cells including duplication CNVs
from three out of the four CMT1A cells examined (Figure 3A).
After customized data processing, the gain in the CMT1A
Figure 3 Array CGH of single cells with submicroscopic genomic imbalances using the NIP2 array. (A) Representative plots of chromosome
views of the specic chromosomes for each studied region adjacent to the enlargement of the boxed region with expected copy number
changes and the anking regions after data analysis with the Agilent Workbench 5.0 using algorithm ADM-2, threshold 5.0. The 6.6 Mb
PWS deletion, the 2.5 Mb DGS deletion, the 1.2 Mb WS deletion, the 3.8 Mb SMS deletion and the reciprocal PTLS duplication, and the
1.3 Mb CMT1A duplication are all detected using single-cell WGA DNA. (B) Chromosome view of array CGH of a single cell with the CMT1A
duplication and the boxed region enlarged below it after customized data analysis. The 1.3 Mb CMT1A gain was apparent; probes with
signicant copy number gain are highlighted in green. The X-axis represents genomic position and the Y-axis shows the normalized log ratios
W. Bi et al.
16
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 7
region was again apparent by visual inspection (Figure 3B).
The expected changes failed to be detected in only one CMT1A
cell. Thus, the detection rate using the NIP2 array was 11/12
single cells analyzed in 13/14 hybridizations performed. The
additional known copy number gains or losses in 5q, 6q, 7q,
9p, 13q, and 14q, in the PWS cell were also detected. For this
array, the autosomal false positive calls were greatly reduced
by ltering out any calls based on 150 probes and an average
log ratio 0.3. Using these settings, there were no more than
ve unexpected calls in 12 out of the 14 hybridizations,
including four hybridizations with no unexpected calls.
Therefore, copy number gains and losses of 1 Mb were
detectable with few false positive calls using the 180 K NIP2
array.
We also evaluated the ability of the NIP2 array to detect
aneuploidy by performing array CGH with WGA DNA from a
single trisomy 21 cell. The gain in chromosome 21 was
detected at a threshold of 3.0, but not at a threshold 5.0. The
known presence of a gain of chromosome 12 in the two
individually analyzed DGS cells was likewise detected at a
threshold of 3.0 but not 5.0. We speculate that the reason for
the relatively poorer performance of this array was related to
the lower density coverage of chromosomes 21 and 12, except
for the subtelomeric regions, which were covered at high
density. This was intentional, as this array was specically
designed to evaluate the most optimal probe selection for
regions of known genomic disorders. We also noted in several
hybridizations that, in general, false-positive copy number
aberration calls were more frequent at the subtelomeric
regions. It is possible that the sequence characteristics of
these regions results in poor performance of many of the
subtelomeric probes; alternatively, WGA of the subtelomeric
DNA may not be as robust or efcient as for other regions in
the genome.
This interpretation, that the detection of aneuploidy is
inuenced by the probe coverage of the specic chromosome
outside the subtelomeric regions, is in part supported by the
better detection of copy number changes of the chromosome
X, as the array had dense probe coverage for six interstitial
regions on X in addition to the subtelomeric regions. The sex
chromosome copy number imbalance was detected for all 14
hybridizations with gender-mismatched reference WGA DNA
at the threshold of 3.0, but two of them were missed at the
threshold of 5.0.
Detection of submicroscopic genomic imbalances using the NIP3
array
The customized NIP3 array has high density probes in the
same targeted regions as the NIP2 array, but the density is even
higher with an average of one probe every 400 bp. Using this
array, we performed single cell array CGH with WGA DNA from
two CMT1A cells and from two PWS cells with gender-
mismatched pooled reference WGA DNA. With a threshold
setting of 5.0, the expected CMT1A gain and PWS deletion
and the changes in the sex chromosomes were detected (data
not shown). After removing copy number aberration calls
detected by 1000 probes or an average log ratio 0.3, all the
known changes were still detected, but there were no false
positive calls for any of the four hybridizations.
A 565 kb duplication of the MECP2 genomic region was not
detected using a single cell
To test the detection limit of single cell array CGH, we
performed array CGH on WGA DNA from single cells having
a 565 kb duplication in Xq28 that includes the MECP2 gene.
WGA DNAs from three single cells were hybridized on the
NIP2 arrays. Gender-matched reference was used for two
hybridizations and gender-mismatched reference was used
for the third hybridization. The expected gain of the MECP2
gene was not revealed for any of the three hybridizations using
a threshold of 5.0 with ltering to remove calls detected by
150 probes or with average log ratio 0.3 (data not shown).
The WGA DNAs from two MECP2-duplication cells were also
tested on the NIP3 array using gender-matched pooled
reference WGA DNA. The MECP2 region was covered by 1334
probes in this array. A copy number gain was detected in
both cells using a threshold of 5.0, but it spanned only part of
the 0.6 Mb duplicated region. We were also unable to nd a
lter setting that effectively removed false positive copy
number aberration calls and still preserved detection of the
duplication.
Thus, a copy number gain of 565 kb on the X chromosome
in a single cell was not consistently detectable using our
customized arrays under the present conditions.
DISCUSSION
In this study we optimized conditions of array CGH using
single cells derived from cell lines. Our data showed that it is
feasible to reliably detect aneuploidies and microdeletions and
microduplications 1 Mb using customized oligonucleotide
arrays. We estimate that the resolution limit using the present
protocol is ~1 Mb because the 1.3 Mb CMT1A duplication and
the 1.2 Mb WS deletion were detected whereas the 565 kb MECP2
duplic ation w as not. This method is potentially useful for
detection of common genomic disorder-associated copy
number changes in preimplantation embryos and using
noninv asively isolated fetal cells. Moreover, this single-cell
analysis can be applied in high resolution copy number
investigations in tumor cells. We showed that it is feasible
to use our customized NI P2 array to de tect aneuploidy
and genomic disorders at a single-cell level. However, to
validate its value in clinical diagnosis, additional tests on
blinded samples are required to determine precisely the
detection sensitivity and specicity.
It is known that the array data plots using WGA DNA from
single cells are much noisier than those from nonamplied
DNA. Following WGA, the quantity of puried DNA obtained
from a single cell is in the range from 2 to 5 μg, about one
High resolution single-cell array CGH
17
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 8
million-fold more than the 6 pg DNA present in a single cell.
Therefore, stochastic variations are unavoidable, which results
in noisier array CGH plots and difculty to reproducibly detect
small copy number changes. In this study, we used the
PicoPlex single cell WGA kit for DNA amplication. This WGA
system has been used to amplify DNA of single cells biopsied
from cleavage stage embryos for the detection of aneuploidy
by array CGH on BAC arrays.
15
To improve the detection
capability of single cell array CGH, we optimized conditions
by determining the most appropriate reference control,
developing customized arrays, and applying specialized data
processing.
By testing different sources of reference DNA for single cell
array CGH, we showed that a hybridization reference control
consisting of WGA product from a single cell resulted in fewer
false positive calls and better sensitivity than using WGA
product from 5 ng of genomic DNA. This suggests that there
is a nonrandom sequence-specic bias with WGA efciency
from single cells, the consequences of which for array-CGH
can be mitigated by using the same source, that is, a single cell,
for the WGA of the test and reference samples. In addition, we
found that pooled reference DNAs containing t he products
of more than ten WGA reacti ons from individual single cells,
originating from one cell culture, further reduced the
number of false positive copy number aberrations. This
indicates that there is also random variability in the
efciency of the W GA from a single cell. By pooling more
than ten WGA reaction products, such random artifacts can
be minimized. An additional advanta ge of using pooled
reference WGA product is that it can be tested for quality
before its usage on precious single-cell test clinical samples.
Therefore, the most suitable reference for single cell array
CGH is pooled WGA product from individually amplied
single cells from one cell cul ture.
The concept that increasing the number of oligonucleotide
probes on arrays improves their sensitivity has been suggested
previously.
10
Using oligo tiling array platforms, these authors
reliably detected 2.6 to 3.0 Mb copy number changes in single
cells. We conrmed this observation by reproducible detection
of the 1.3 Mb CMT1A duplication using an array (NIP1) with
20,000 probes in that region. To extend the high sensitivity
detection to all common known genomic disorders, we
designed a customized NIP2 array with a high density of
probes covering the relevant genomic regions, while the other
regions in the genome are covered by less densely spaced
probes. In addition to the increased sensitivity in the targeted
regions, the advantage of this design is that the false positive
calls in irrelevant regions could be easily removed by a
probe-number lter. The detection of aneuploidy using the
NIP2 array was not optimal, but can be improved by increasing
probe coverage in the nonsubtelomeric regions and removing
poorly performing probes in the subtelomeric regions. This
should also improve the detection of copy number changes
in the subtelomeric regions.
The false positive calls in single-cell array CGH are likely due
to unfaithful amplications during WGA or tissue culture
artifacts. It may be difcult to determine whether a detected
copy number change is a true one that is present in the cell,
or a false positive call. True changes are more likely to be
present reproducibly in independent hybridizations with
WGA DNA from different single cells of the same source. For
example, we compared the copy number changes detected
by array CGH using the NIP2 array in two individually treated
single cells from a cell line established from a single patient
having a CMT1A duplication. Using threshold 3.0 and a lter
to remove changes detected by 30 probes or by an absolute
average log ratio of 0.3, the unexpected autosomal copy
number changes were 42 and 43, respectively. Comparison of
the detections in these two individual cells showed that the
only 1 Mb change shared by these two cells was the expected
CMT1A gain. Therefore, when multiple cells from one cell
source are available, the true copy number changes may be
differentiated from the false positive detections by comparison
of array CGH results in multiple individual cells.
This method of single-cell array CGH is potentially useful for
preimplantation genetic diagnosis to extend the detection
ability from the current screening for unbalanced translocation
derivatives and aneuploidy to detection of copy number
aberrations in regions relevant to common genomic disorders.
16
This method is also potentially applicable for noninvasive
prenatal genetic diagnosis using intact circulating fetal cells
and m eets the challenge that very few fetal cells will be
available for analysis.
Both intact fetal cells and cff nucleic acids in the maternal
circulation have been explored for noninvasive prenatal
genetic diagnosis.
17,18
Rapid progress has been made on the
application of massively parallel sequencing technology on cff
DNA. A recent study demonstrated that ~11% of DNA sequence
reads from maternal plasma were of fetal origin and indicated
that the fetal genome is relatively uniformly represented in cff
DNA,
19
although there could still be underrepresentation
of certain sequences compared to DNA originating from
intact fetal cells. Furthermore, to date, sequencing of
maternal plasma DNA mainly focused on fetal aneuploidy
detection,
20,21
and further studies are necessary to investigate
whether it is feasible to use deep sequencing to detect
microdeletions or microduplications that can be associated with
severe clinical consequences. In addition, although the cost per
nucleotide of DNA sequencing has diminished dramatically in
recent years, the current cost for deep sequencing is still high
and the entire process, including data analysis, is time-
consuming. In contrast, the application of more cost-efcient
array technology to cff DNA is limited by the current inability
to efciently isolate free fetal from free maternal DNA. Thus,
intact fetal cells retain these important potential advantages
for noninvasive prenatal diagnosis. While we show here that
copy number aberrations as small as 1 Mb are potentially
detectable through analysis of single fetal cells retrieved
W. Bi et al.
18
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 9
noninvasively during pregnancy, the major challenge to con-
sistently isolate the extremely r are fetal cells remains. Work
by us and others is in progress to develop new meth ods to
label cells using fetal-specic markers and separate them
from maternal cells using microuidics and other advanced
cell-sorting technologies. In addition, fetal cells have been
identied in maternal endocervix in a much higher percentage
than in maternal circulation, which provides an additional
potential source for isolation of fetal cells.
2,3
One concern for
prenatal diagnosis using fetal cells is that some types of
fetal cells (e.g. CD34-positive cells) can persist in the
maternal circulation for a long time.
22
Therefore, isolated
fetal cells may not be derived from the ongoing pregnancy,
but from a previous one. However, in contrast to older
single-cell uoresce nce in situ hybridization (FISH)-based
approaches,
17
array-based copy number analysis requires
prior amplication of the single-cells genome. Usually,
more WGA DNA is obtained than needed for array-analysis
andsomeWGADNAcanbeusedtoconrm fetal origin of
the isolated cells by polymorphic marker analysis. By
comparing the genotypes of isolated fetal cells to that of
parents, siblings, or prior last pregnancies (if material is
available), fetal allele information of the current pregnancy
can be obtained a nd thi s prob lem can b e resolve d.
In summary, this study shows that it is feasible to detect
aneuploidy and microdeletions and microduplications of
1 Mb in a single cell. The methods we developed can
potentially be utilized for preimplantation genetic diagnosis
of fetal aneuploidy and common genomic disorders, for similar
purposes in noninvasive prenatal diagnosis, and for studies of
tumor heterogeneity and micrometastasis.
ACKNOWLEDGEMENTS
We are grateful to Ankita Patel, Christine Eng, Joanna
Wiszniewska, and Joel M. Sederstrom for helpful advices. We
would like to thank Lance Cooper, Zhishuo Ou, Patricia Eng,
and Patricia Hixson for excellent technical assistance and
Marjorie A. Withers for expert assistance with cell cultures.
WHATS ALREADY KNOWN ABOUT THIS TOPIC?
Single-cell array CGH has been applied in PGD for a compre-
hensive chromosome analysis including aneuploidy analysis of all
24 chromosomes.
Single cell technol ogy has the potential to be applied in
noninvasive prenatal diagnosis using isolated intact fetal cells
and meets the challenge that very few fetal cells will be availa ble
for analysis.
The current resolution of single-cell array CGH is not high
enough for detection of most microdeletion or microduplication
syndromes.
WHAT DOES THIS STUDY ADD?
We develop and optimize methods to reliably detect genomic
imbalances of > 1 Mb by array CGH using single cells.
This method of single-cell a rray CGH is potentially useful for
PGD and noninvasive prenatal genetic diagnosis to detect copy
number aberrations in regions relevant to common genomic
disorders.
REFERENCES
1. Lo YM, Lo ES, Watson N, et al. Two-way cell trafc between mother and
fetus: biologic and clinic al implicat ions. Blood 1996; 88: 43905.
2. Adinol M, Sherlock J. Fetal cells in transcervical samples at an early
stage of gestation. J Hum Genet 2001; 46:99104.
3. Imudia AN, Kumar S, Diamond MP, DeCherney AH, Armant DR
Transcervical retrieval of fetal cells in the practice of modern medicine:
a review of the current literature and future direction. Fertil Steril 2010;
93: 172530.
4. Bianchi DW, Williams JM, Sullivan LM, Hanson FW, Klinger KW, Shuber
AP. PCR quantitation of fetal cells in maternal blood in normal and
aneuploid pregnancies. Am J Hum Genet 1997; 61: 8229.
5. Wang D, Bodovitz S. Single cell analysis: the new frontier in omics.
Trends Biotechnol 2010; 28: 28190.
6. Van den Veyver IB, Patel A, Shaw CA, et al.Clinicaluseofarray
comparative genomic hybridization (aCGH) for prenatal diagnosis
in 300 cases. Prenat Diagn 2009; 29:2939.
7. Hillman SC, Pretlove S, Coomarasamy A, et al. Additional information
from array comparative genomic hybridization technology over
conventional karyotyping in prenatal diagnosis: a systematic review and
meta-analysis. Ultrasound Obstet Gynecol 2011; 37:614.
8. SpitsC,LeCaignecC,DeRyckeM,et al . Whole-genome multiple
displacement amplication from single cells. Nat Protoc 2006; 1:196570.
9. Fiegler H, Geigl JB, Langer S, et al. High resolution array-CGH analysis of
single cells. Nucleic Acids Res 2007; 35: e15.
10. Geigl JB, Obenauf AC, Waldispuehl-Geigl J, et al. Identication of small
gains and losses in single cells after whole genome ampli
cation on
tiling oligo arrays. Nucleic Acids Res 2009; 37: e105.
11. Stankiewicz P, Lupski JR. Structural variation in the human genome and
its role in disease. Annu Rev Med 2010; 61: 43755.
12. Bi W, Sapir T, Shchelochkov OA, et al. Increased LIS1 expression affects
human and mouse brain development. Nat Genet 2009; 41: 16877.
13. Carvalho CM, Zhang F, Liu P, et al. Complex rearrangements in patients
with duplications of MECP2 can occur by fork stalling and template
switching. Hum Mol Genet 2009; 18: 2188203.
14. Ou Z, Kang SH, Shaw CA, et al. Bacterial articial chromosome-
emulation oligonucleotide arrays for targeted clinical array-
comparative genomic hybridization analys es. Genet Med 2008; 10:
27889.
15. Gutierrez-Mateo C, Colls P, Sanchez-Garcia J, et al. Validation of
microarray comparative genomic hybridization for comprehensive
chromosome analysis of embryos. Fertil Steril 2010. (Epub ahead of
print).
16. Harper JC, Harton G. The use of arrays in preimplantation genetic
diagnosis and screening. Fertil Steril 2010; 94: 11737.
High resolution single-cell array CGH
19
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 10
17. Bianchi DW, Simpson JL, Jackson LG, et al. Fetal gender and aneuploidy
detection using fetal cells in maternal blood: analysis of NIFTY I data.
National Institute of Child Health and Development Fetal Cell Isolation
Study. Prenat Diagn 2002; 22: 60915.
18. Beaudet AL. Progress toward Noninvasive Prenatal Diagnosis. Clin
Chem 2011; 57 : 8024.
19. Lo YM, Chan KC, Sun H, et al. Maternal plasma DNA sequencing reveals
the genome-wide genetic and mutational prole of the fetus. Sci Transl
Med 2011; 2: 61ra91.
20. Chiu RW, Akolekar R, Zheng YW, et al. Non -invasiv e prenatal
assessment of trisomy 21 by multiplexed maternal plasma DNA
sequencing: large scale validity study. BMJ 2011; 342: c7401.
21. Ehrich M, Deciu C, Zwiefelhofer T, et al. Noninvasive detection of fetal
trisomy 21 by sequencing of DNA in maternal blood: a study in a clinical
setting. Am J Obstet Gynecol 2011; 204(205): e1e11.
22. Bianchi DW, Zickwolf GK, Weil GJ, Sylvester S, DeMaria MA. Male
fetal progenitor cells persist in maternal blo od for as lo ng as 27 years
postpartum. Proc Natl Acad Sci USA 1996; 93: 7058.
W. Bi et al.
20
Prenatal Diagnosis 2012, 32,1020 © 2012 John Wiley & Sons, Ltd.
Page 11
  • Source
    • "Next-generation whole-genome DNA sequencing (NGS or DNA-Seq) has also provided an additional platform besides qPCR and array-based methods to assess aneuploidy comprehensively in human embryos, including at the single-cell level (Hou et al. 2013; Fiorentino et al. 2014; Wang et al. 2014). Since aCGH covers approximately 30 % of the genome and can only detect gross chromosomal abnormalities at a lower threshold of approximately 1 MB (Bi et al. 2012), sub-chromosomal aberrations that are smaller than this resolution and missed by aCGH would be detectable by NGS in theory. In addition to examining copy number variation at a single nucleotide level, DNA-seq is also more cost-effective than aCGH since 48-96 samples or more can be simultaneously sequenced with high coverage by multiplexing with barcoded adapters. "
    [Show abstract] [Hide abstract] ABSTRACT: Formation of a totipotent blastocyst capable of implantation is one of the first major milestones in early mammalian embryogenesis, but less than half of in vitro fertilized embryos from most mammals will progress to this stage of development. Whole chromosomal abnormalities, or aneuploidy, are key determinants of whether human embryos will arrest or reach the blastocyst stage. Depending on the type of chromosomal abnormality, however, certain embryos still form blastocysts and may be morphologically indistinguishable from chromosomally normal embryos. Despite the implementation of pre-implantation genetic screening and other advanced in vitro fertilization (IVF) techniques, the identification of aneuploid embryos remains complicated by high rates of mosaicism, atypical cell division, cellular fragmentation, sub-chromosomal instability, and micro-/multi-nucleation. Moreover, several of these processes occur in vivo following natural human conception, suggesting that they are not simply a consequence of culture conditions. Recent technological achievements in genetic, epigenetic, chromosomal, and non-invasive imaging have provided additional embryo assessment approaches, particularly at the single-cell level, and clinical trials investigating their efficacy are continuing to emerge. In this review, we summarize the potential mechanisms by which aneuploidy may arise, the various detection methods, and the technical advances (such as time-lapse imaging, "-omic" profiling, and next-generation sequencing) that have assisted in obtaining this data. We also discuss the possibility of aneuploidy resolution in embryos via various corrective mechanisms, including multi-polar divisions, fragment resorption, endoreduplication, and blastomere exclusion, and conclude by examining the potential implications of these findings for IVF success and human fecundity.
    Full-text · Article · Nov 2015 · Cell and Tissue Research
  • Source
    • "Standard DNA-microarrays can detect copy number variations (CNVs) larger than 2.5 Mb from a single-cell genome [20]–[22], while targeted array comparative genomic hybridizations can discover approximately 1 Mb-sized DNA imbalances [23], although remarkably, CNVs as small as 56 kb in single-cell PCR-based WGA products have been detected [24]. Similarly, SNP-arrays can find copy number aberrations encompassing millions of bases in a cell [25]–[28], but have the advantage of enabling the discovery of copy neutral DNA anomalies and regions of loss-of-heterozygosity (LOH), and allow inferring genome-wide haplotypes [29]–[31]. "
    [Show abstract] [Hide abstract] ABSTRACT: Advances in whole-genome and whole-transcriptome amplification have permitted the sequencing of the minute amounts of DNA and RNA present in a single cell, offering a window into the extent and nature of genomic and transcriptomic heterogeneity which occurs in both normal development and disease. Single-cell approaches stand poised to revolutionise our capacity to understand the scale of genomic, epigenomic, and transcriptomic diversity that occurs during the lifetime of an individual organism. Here, we review the major technological and biological breakthroughs achieved, describe the remaining challenges to overcome, and provide a glimpse into the promise of recent and future developments.
    Full-text · Article · Jan 2014 · PLoS Genetics
  • Source
    • "The third major improvement could be achieved by usage of high-resolution aCGH arrays based on oligonucleotides. Studies published in the recent years indicated applicability of the new generation of high-resolution SNP and CGH arrays for single-cell CGH analysis [26], [35], [39]–[43]. In comparison to the previously used BAC-based arrays [24], [25], [44] these technologies offer lower qualitative variability of the array slide manufacturing process, customizable microarray designs and very low probe spacing. "
    [Show abstract] [Hide abstract] ABSTRACT: Disseminated cancer cells (DCCs) and circulating tumor cells (CTCs) are extremely rare, but comprise the precursors cells of distant metastases or therapy resistant cells. The detailed molecular analysis of these cells may help to identify key events of cancer cell dissemination, metastatic colony formation and systemic therapy escape. Using the Ampli1™ whole genome amplification (WGA) technology and high-resolution oligonucleotide aCGH microarrays we optimized conditions for the analysis of structural copy number changes. The protocol presented here enables reliable detection of numerical genomic alterations as small as 0.1 Mb in a single cell. Analysis of single cells from well-characterized cell lines and single normal cells confirmed the stringent quantitative nature of the amplification and hybridization protocol. Importantly, fixation and staining procedures used to detect DCCs showed no significant impact on the outcome of the analysis, proving the clinical usability of our method. In a proof-of-principle study we tracked the chromosomal changes of single DCCs over a full course of high-dose chemotherapy treatment by isolating and analyzing DCCs of an individual breast cancer patient at four different time points. The protocol enables detailed genome analysis of DCCs and thereby assessment of the clonal evolution during the natural course of the disease and under selection pressures. The results from an exemplary patient provide evidence that DCCs surviving selective therapeutic conditions may be recruited from a pool of genomically less advanced cells, which display a stable subset of specific genomic alterations.
    Full-text · Article · Jan 2014 · PLoS ONE
Show more