Tumor progression of culture-adapted human embryonic stem cells during long-term culture

Tumor Progression of Culture-Adapted HumanEmbryonic Stem Cells During Long-Term Culture

Sheng Yang,1,†,‡ Ge Lin,1,† Yue-Qiu Tan,1 Di Zhou,1 Lei-Yu Deng,1 De-Hua Cheng,1 Shu-Wei Luo,1 Tian-Cheng Liu,1

Xiao-Ying Zhou,1 Zheng Sun,1 Yang Xiang,1 Tian-Ji Chen,1 Ji-Fang Wen,2 and Guang-Xiu Lu1*

1National Centerof Human StemCell Research and Engineering,Institute of Human Reproduction & StemCell Engineering,Central South University,Changsha,China2Departmentof Pathology,Basic Medical College,Central South University,Changsha,China

Human embryonic stem cells (hESCs) during long-term culture acquire chromosomal changes similar to those occurring in tu-

morigenesis. This was raised concerns about the progression from hESCs to malignant cells. This study aimed to investigate

the changes in chromosomes, cell phenotype, and genes in culture-adapted hESCs to ascertain whether tumorigenic transfor-

mation occurred. By cytogenetic analysis we found progressive karyotypic changes from simple to complex in chHES-3, one of

the hESC lines established in our laboratory, during a long-term suboptimal culture. We further compared chHES-3 cells at dif-

ferent karyotypic stages in cell surface markers, in vivo differentiation, cell cycle, apoptosis, and gene expression profiles. We

found that the karyotypically aberrant chHES-3 had higher S-phase fraction in cell cycle distributions and antiapoptosis ability.

In vivo differentiation of karyotypically normal chHES-3 resulted in relatively mature teratoma, whereas karyotypically aberrant

chHES-3 formed immature teratoma (grade III), in which more primary neural epithelium was revealed by pathological analysis.

The microarray analysis and real-time PCR results showed that some oncogenes were upregulated in karyotypically aberrant

chHES-3 cells, whereas the genes related to differentiation were downregulated, and that Wnt signal pathway was activated. In

conclusion, chHES-3 cells underwent deregulation of self-renewal and dysfunction of related genes in long-term culture adapta-

tion, leading to malignant transformation. VVC 2008 Wiley-Liss, Inc.

INTRODUCTION

Since the first human embryonic stem cell

(hESC) line was established (Thomson et al., 1998),

hESCs have held great promise for future cell- and

tissue-replacement therapy (Cowan et al., 2004).

However, concerns have been raised on the safety of

hESCs in long-term culture. Andrews et al., (2005)

reported that, after multiple passaging, hESCs

acquired chromosomal aberrations similar to those

found in embryonal carcinoma (EC) cells (Draper

et al., 2004). Several recent reports were provided

further evidence that hESCs acquire chromosomal

abnormalities when cultured in suboptimal culture

conditions, such as the presence of trypsin in passag-

ing (Cowan et al., 2004; Inzunza et al., 2004;

Zeng et al., 2004;Hoffman andCarpenter, 2005;Mita-

lipova et al., 2005; Baker et al., 2006; Imreh et al.,

2006). It has also been reported that the karyotypic

changes might lead to tumorigenesis (Enver et al.,

2005; Andrews, 2006; Baker et al., 2006). But until

now, karyotypic changes in a single subline of hESCs

has not been studied in detail. Also, the molecular

mechanisms of cellular malignant transformation of

culture-adapted hESCs remain poorly understood.

Andrews (2006) suggested that the karyotypi-

cally aberrant hESCs were likely to tend towards

the transformed phenotype of the tumor stem

cells. But until now, the relationship between kar-

yotypic changes and the malignant transformation

of hESCs has not been studied in detail. In

this study, we focused on whether karyotypic

changes of hESCs lead to cellular malignant trans-

formation or tumorigenesis, investigating cytoge-

netic changes, cellular biological characteristics,

gene expression profiles, and in vivo teratoma for-

mation of the karyotypically aberrant hESCs at

the different stages. We found that the gradual

†Present address: The 2nd Affiliated Hospital of WenZhou Medi-cal College, Wenzhou, China.Received 20 September 2007; Accepted 14 April 2008

DOI 10.1002/gcc.20574

Published online 9 May 2008 inWiley InterScience (www.interscience.wiley.com).

†Sheng Yang and Ge Lin contributed equally to this work.*Correspondence to: Guang-Xiu Lu, National Center of Human

Stem Cell Research and Engineering, Institute of Human Repro-duction and Stem Cell Engineering, Central South University,Changsha, China. E-mail: [email protected]

Supported by: The Hi-Tech Research And Development ofChina, Grant number: 2006 AA02A102; The National BasicResearch Program of China, Grant number: 973 program No. 00CB51010; Bureau Science and Technology Key Project Funds ofHunan Province, Grant number: 03SSY2001; Ministry of Educationof China (Research Fund for the Doctoral Program), Grant number:20030533002; The Key Project of Chinese Ministry of Education,Grant number: 105131.

VVC 2008 Wiley-Liss, Inc.

GENES, CHROMOSOMES & CANCER 47:665–679 (2008)

karyotype changes occurred in chHES-3 cells dur-

ing long-term culture and that the aberrant chHES-

3 cells underwent malignant transformation, as

well as that the activation of Wnt signaling path-

way might be the main reason for the cellular ma-

lignant transformation and tumorigenesis.

MATERIALS ANDMETHODS

Derivation and Culturing of chHES-3

chHES-3, one of hESC lines established in our

laboratory, was isolated by immunosurgery and cul-

tured in serum-free KSR media composed of

Knock-out DMEM (Gibco BRL, Gaithersburg,

MD), supplemented with 15% serum replacement

(Gibco BRL, Gaithersburg, MD), 0.1 mM b-mer-

captoethanol (Sigma, Chemical), 1% nonessential

amino acids (Gibco-BRL, St. Louis, MO), 2 mM

L-glutamine (Gibco BRL, Grand Island), 50 U/ml

penicillin (Sigma, St Louis, MO), 50 lg/ml strepto-

mycin (Sigma, St Louis, MO), and 4 ng/ml human

recombinant basic fibroblast growth factor (Gibco

BRL, Gaithersburg, MD). The chHES-3 cells were

initially propagated on mouse embryonic fibro-

blasts (mEFs) at higher density (4–5 3 104 cells/

cm2), which were derived from inbred Kunming

White mice at 12.5 d of gestation and were inacti-

vated by mitomycin-C (10 lg/ml) (Xie et al., 2004).

The chHES-3 cells were passaged by mechanical

dissection or by collagenase IV digestion (200 U/

ml) (Sigma, St. Louis, MO) followed by mechani-

cal slicing for every 6–7 days. Thawed chHES-3

cells from early passages were cultured on either

higher-density feeders mentioned earlier or lower-

density feeders (2 3 104 cells/cm2). The chHES-3

cells were cultured on 6-well tissue culture plates

in clone-like clumps, picked out the undifferentia-

tion ES cells under inversion microscope, and were

detected by ES markers, including SSEA-4, before

any analysis. The hES cell lines used in this study

were derived from surplus embryos from IVF treat-

ment with informed consent of the patients.

Karyotype Analysis and FISH

Karyotype analysis by G-banding was performed

with about 50 metaphases for every several pas-

sages, and FISH was performed using commer-

cially available whole chromosome painting (WCP)

probes, including WCP1, WCP4, WCP6, WCP7,

and WCP8 (gift from Prof. Xinyuan Guang, The

University of Hong Kong).

CGH Analysis

CGH analysis was performed using the protocol

developed by Klein and Schmidt-Kittler (1999). In

brief, �5 3 105 karyotypically aberrant chESC-3cells were isolated and then lysed by Proteinase K.

Their genomic DNA was digested into appropriate

fragments of 300–800 bp, which were then labeled

with Spectrum Green (Vysis, USA) as probes. The

reference genomic DNA fragments (from normal

karyotypic human blood cells) was labeled with

Spectrum Red (Vysis, USA) as probes. The two

kinds of probes were co-hybridized to metaphase

slides. After hybridization in a moist chamber for

72 hr at 378C, the slides were washed and then an-

alyzed using Quips CGH karyotyping Imaging

Software (Vysis, USA). Ten to 15 metaphases were

captured and evaluated for each sample. Red/green

ratios of 1:�1.25 indicated amplified regions and

ratios of 1:�0.75 indicated deleted regions.

RT-PCR

Total RNA was isolated from cell pellets using

TRIZ reagent (Sigma, USA) and reversely tran-

scribed in a thermocycler using virus reverse tran-

scriptase and random primers, according to the

manufacturer’s protocol supplied by Fermentas

Company (Ferments, Lithuania). PCR amplifica-

tions of different genes were performed using Taq

DNA polymerase (Promega Madison, WI). The

PCR products were subjected to electrophoresis in

1.5% agarose gel containing ethidium bromide

(0.5 lg/ml) with PUC Mix marker 8 as marker.

(MBI-Fermentas, St. Leon-Roth. Nukleotide).

The primers for RT-PCR were listed in Table 1.

Real-Time RT-PCR Analysis

Total RNA was extracted from 1 3 106 hESCs

by classical method and then digested by the

DNase I (NEB, Frankfurt a.M., Germany). About

4 ll aliquots of total RNA were reversely tran-

scribed in a T-GRADIENT thermocycler (Bio-

metra, Gottingen) by using the RevertAidTM First

Strand cDNA Synthesis Kit (Fermentas, Lithua-

nia). Real-time RT-PCR was performed by using

the Lightcycler Faststart master DNA Sybr Green

I kit (Roche, Germany) with specific primer pairs

listed in Table 2. A melting curve analysis was per-

formed to determine the specificity of the ampli-

fied products. The relative gene expression level

was calculated by subtracting the cycle threshold

(Ct) value of the 28 s (control gene) from Ct valueof the target gene to generate the DCt value, andthe relative changes in mRNA expression of aber-

Genes, Chromosomes & Cancer DOI 10.1002/gcc

666 YANG ETAL.

rant hESCs versus normal hESCs were obtained

using the 2-DDCt method (Livak and Schmittgen,

2001) and 95% confidence interval was accepted.

Telomerase Activity

Telomerase activity was measured using TRA-

Peze Telomerase detection kit (Chemicon, Teme-

cula, CA) according to the manufacturer’s instruc-

tions as described (Kim et al., 1994; Weinrich et al.,

1997). In brief, after addition of 10–20 ll telomerase

assay lysis buffer (13 CHAPS), the cells were lysed

on ice. The lysate was incubated on ice for 30 min

and then centrifuged at 12,000 rpm for 20 min at

48C. The supernatant was collected and the protein

concentration was determined by standard proce-

dures (BCA protein assay). A volume of 0.33 lg pro-

tein equivalent was added to a 48 ll reaction solu-

tion consisting of TRAP buffer, dNTP Mix, TS

primer, RP primer mix, and 2 U Taq polymerase.

The PCR condition was 33 cycles of 948C for 30 sec

and 598C for 30 sec. The PCR products were deter-

mined by electrophoresis in 12.5% nondenaturing

polyacrylamide gel with silver staining.

Detection of HESC-Specific Markers

Cells were fixed in 100% ethanol, washed with

13 phosphate buffered solution (13 PBS), and

incubated with monoclonal primary antibodies

against SSEA-1, SSEA-4, TRA-1-81, and TRA-1-

60, separately. Then, the cells were stained by

using rabbit anti-mouse immunoglobulins conju-

gated to fluorescein isothiocyanate (Chemicon,

Temecula, CA). Oct-4 staining was performed

according to the manufacturer’s instructions (Santa

Cruz Biotechnology, Santa Cruz, CA). Alkaline

phosphatase activity was detected according to the

protocol of the Fast Red Substrate Pack (Invi-

trogen, Germany).

TABLE 1. Primers Used for RT-PCR

Gene Forward primer (50-30) Reverse primer (50-30) Tm

Size ofproduct

FGFR4 GTTTCCCCTATGTGCAAGTCC GCGCTGCTGCGGTCCATGT 57 194TERF1 GCAACAGCGCAGAGGCTATTATT AGGGCTGATTCCAAGGGTGTAA 57 159TERF2 AAACGAAAGTTCAGCCCCG TCCTCCAAGACCAATCTGCTTA 55 92OCT3/4 CTTGCTGCAGAAGTGGGTGGAGGAA CTGCAGTGTGGGTTTCGGGCA 64 168REX-1 TGAAAGCCCACATCCTAACG CAAGCTATCCTCCTGCTTTGG 57 556THY1 TCTCAGGGACTTCTGCGGG GGTTGGGAAAAGCCATTTCC 57 88LEFTYA GGGAATTGGGATACCTGGAT CTAAATATGCACGGGCAAGG 57 207SOX2 AGT CTC CAA GCG ACG AAA AA GCA AGA AGC CTC TCC TTG AA 54 142NANOG ACTGTCTCTCCTCTTCCCTCCTCC GTAGAGGCTGGGGTAGGTAGGTG 64 387cripto-1 TCCTTCTACGGACGGAACTG AGAAATGCCTGAGGAAAGCA 58 139

TABLE 2. Primers Used for Real-Time RT-PCR

Gene Forward primer 50-30 Reverse primer 50-30 ImageID Ta

Nanoga CCTGTGATTTGTGGGCCTG GACAGTCTCCGTGTGAGGCAT NM_024865 60POU5F1a GTGGAGGAAGCTGACAACAA ATTCTCCAGGTTGCCTCTCA NM_002701 60Sox2a GTATCAGGAGTTGTCAAGGCAGAG TCCTAGTCTTAAAGAGGCAGCAAAC NM_003106 60BuB1 ATGTTGAGCAGGTTGTTA TATGTTAGTTAGTTGCCTCTTTC NM_004336 50TERF1 TTGCCAGTTGAGAACGATA GGGCTGATTCCAAGGGTGT NM_017489 56GSK-3b CTGTTCCGAAGTTTAGCC AAGAGGTTCTGCGGTTTA NM_002093 50b-Catenin CTGCCAAGTGGGTGGTATA GGGATGGTGGGTGTAAGAG NM_001904 54Cyclin D1b GATCAAGTGTGACCCGGACT TCCTCCTCCTCTTCCTCCTC NM_053056 56FGFR1 GGACAAGGACAAACCCAA GCGTCCGACTTCAACAT NM_000604 52CDK4 GAAACTCTGAAGCCGACCAG ACATCTCGAGGCCAGTCATC NM_053056 58V-jun ACTCGGACCTCCTCACCTCG ATGTGCCCGTTGCTGGAC NM_002228 61Tp53 CCACCATCCACTACAACTACAT AAACACGCACCTCAAAGC NM_000546 54NF1 GCTGAAAGCACCAAACG CTGCCTACTTCCTCCAT NM_000267 50AFP GGGAGCGGCTGACATTAT TGTTTCATCCACCACCAA NM_001134 53PAX6a CCAGGGCAATCGGTGGTAGT ACGGGCACTCCCGCTTATAC NM_000280 60EOMESa CGGCCTCTGTGGCTCAAA AAGGAAACATGCGCCTGC NM_005442 6028Sb GAACTTTGAAGGCCGAAGTG ATCTGAACCCGACTCCCTTT 60

aThese primers were designed as described previously (Babaie et al., 2007).bThese primers were designed as described previously (Becker et al., 2006).


667TUMOR PROGRESSION OF HUMAN EMBRYONIC STEM CELL

Teratoma Formation and Analysis

The hESCs were mechanically cut into small

pieces (�50–100 cells per piece) and about 0.5–1

3 106 cells were injected into the rear leg of 6- to

8-weeks-old NOD/SCID mice. Ten weeks later,

the mice were sacrificed and the xenografts were

removed followed by fixation in 4% paraformalde-

hyde (Sigma, St. Louis, MO) for 24 hr. After paraf-

fin embedment, the tumors were sectioned and

stained by haematoxylin and eosin (H&E). In the

experiments of the differentiation in vivo, the

SCID mice care was in accordance with guidelines

for the Care and Use of Laboratory Animals, enun-

ciated by the Ministry of Science and Technology

of the People’s Republic of China.

Cell Proliferation Analysis

For analysis of the cell cycle, 1 3 107 ES

cells were cultured for 12 hr in 20 lm EdU (5-

ethynyl-20-deoxyuridine) medium and then har-

vested and stained by The Click-iTTM EdU Alexa

Fluor1 488 Cell Proliferation Assay Kit (Invi-

trogen, USA) in accordance with the manufac-

turer’s protocol. Fluorescence data were collected

using a FACScalibur (Becton Dickinson, Mountain

View, CA).

Differentiation Experiments In Vitro

For in vitro differentiation, chHES-3 colonies

cultured on mouse embryonic feeder layer were

mechanically cut into small pieces of 50–100 cells

and continued to grow as aggregates in suspension

in ES medium without bFGF for 10 days to form

embryoid bodies. Embryoid bodies were then

transferred onto gelatin-coated 6-well plate for ad-

herent culture for 15 days in the same medium, fol-

lowed by immunocytochemistry analysis. Further-

more, the in vitro differentiation of chHES-3 cells

was examined in routine culture.

Apoptosis Analysis

To measure early apoptosis, cell apoptosis was

examined using Annexin V-FITC kit (Bender

MedSystems, Vienna, Austria). Fluorescence data

were collected using a FACScalibur (Becton Dick-

inson, Mountain View, CA) and further analyzed

using the ModFit 2.0 software (Becton Dickinson,

Mountain View, CA). The apoptotic resistance

experiments were performed by using mitomycin

C. The chHES-3 cells were cultured in 0.01 lg/ml

mitomycin C for 10 hr and then harvested to detect

cell apoptosis as mentioned earlier.

Microarrary and Data Analysis

The normal chHES-3 cells (P30) and aberrant

chHES-3 cells with a simple duplication karyotype

(SIMP) of passage 72 (P72) and complex karyotype

(COMP) (P182) were collected for gene expression

profiles. Total RNA was extracted from normal and

karyotypically aberrant chHES-3 cells using the

RNeasy kit (Qiagen, Chatsworth, CA). One micro-

gram of total RNAwas primed with 100 ng of Oligo

dT-T7 primer and reverse transcribed with Super-

script II (Invitrogen, USA). A second strand was

synthesized and the double strand cDNAwas puri-

fied with DNA Clean and Concentrator (Zymo

Research, Orange, CA). The in vitro transcription

reactions were performed with T7 RNA polymer-

ase. The amplified RNA (aRNA) in the first round

amplification was purified with the RNeasy Mini

Kit (Qiagen, Valencia, CA). The second round

amplification was performed similarly to the first

round, but with 100 ng of aRNA and 500 ng ran-

dom hexamers. The ENZO BioArray HighYield

RNA Transcript Labeling Kit (Affymetrix, USA)

was used to incorporate biotin-labeled nucleotides

in the second round dscDNA, after which RNA

was purified using RNeasy. Fragmentation was

completed using a standard protocol. Prior to

hybridization on Gene Chip array, a test3 array of

housekeeping controls was analyzed to determine

sample suitability for Gene Chip arrays. Hybri-

dized arrays were subsequently scanned for data

analysis. A detailed RNA amplification protocol is

available upon request. The hybridization mixture

was heated at 998C for 5 min and then at 458C for

5 min, followed by centrifugation at 13,000g for

5 min. Gene chips were prehybridized in 200 ml of

13 hybridization buffer for 10 min at 458C with

mixing at 60 rpm in the hybridization oven. The

prehybridization buffer was replaced with 200 ml

hybridization mixture and incubated for 16 hr at

458C, mixed at 60 rpm. The hybridization mixture

was removed and stored at 2708C. Each chip was

filled with 250 ml of nonstringent wash buffer (63SSPE, 0.01% Tween-20). The chips were scanned

with an Affymetrix Scanner 3000 (Affymetrix), and

the gene expression signal was collected using

Affymetrix GCOS V1.1.1 software. The distribu-

tion of upregulated and downregulated genes on

each chromosome was analyzed by the Dchip2004

software. Hierarchical cluster analysis was per-

formed using the AVADIS software.


668 YANG ETAL.

RESULTS

Characterization of the chHES-3 Cell Line

The chHES-3 cell line used in these studies was

established in our laboratory, and was cultured on

feeder cells at higher density than normal ES cells

(see Material and Methods). The cell immunohis-

tochemistry always showed that the chHES-3 cells

are positive for SSEA-3, SSEA-4, TRA-1-60, TRA-

1-81, and oct-4, and negative for SSEA-1 and that

the cells had alkaline phosphatase activity and

telomerase. The results of RT-PCR indicated that

the cells expressed special markers, such as Oct-4,

Nanog, Sox-2, TERF1, TERF2, REX, FGF4,

Cripto, Thy1, and LEFTYA. The chHES-3 cells

presented a normal karyotype and could form tera-

toma after injection into SCID mice, which con-

tained many kinds of cells derived from three

germ layers (see later). All the above mentioned

results confirmed that the chHES-3 cells had basic

features of the normal HES cells. The chHES-3

cell lines kept these basic features of hESCs even-

though their karyotype and characteristics of tera-

toma changed in later passages during long-term

culture (Figs. 1A–1C).

Gradually Karyotypic Changes of the chHES-3 Cell

Line During Long-Term Culture

To monitor the genetic stability of hESCs, we

routinely examined karyotype of hESCs every 5–

10 passages, and found that the karyotype of

chHES-3 cell line displayed a slowly progressive

changes. The G-banding results showed that the

chHES-3 cell line retained a normal karyotype up

to P27 (Fig. 2A); at P34, all analyzed 50 meta-

phases had the same karyotype of 46,XX,dup(1)

(p32p36) (Fig. 2B); from P44 to P99, more chromo-

somal abnormalities accumulated and showed a

karyotype of 46,XX,dup(1)(p32p36), t(1;6;4)(q25;

q23;p16),ins(4;1)(p16;q21q25), t(7;8) (q32;q13)

(Fig. 2C); at P114, a mosaic karyotype was

observed, including the chromosomal abnormal-

ities mentioned above and three kinds of addi-

tional chromosomal abnormalities (Figs. 2D–2F);

from P142 to P153, the cells only demonstrated the

Figure 1. Identical morphology of normal chHES-3, simple duplication chHES-3 (SIMP), and karyotypicallycomplex chHES-3 (COMP) colonies. (A) All colonies stained positive for AKP, SSEA-4, OCT-4, TRA-1-60, andTRA-1-81 and negative for SSEA-1. (B) RT-PCR detection of pluripotency-related genes showed that thesegenes were all expressed in normal, SIMP and COMP chHES-3 cells. (C) The telomerase activity was presentin normal chHES-3 cells (Lane 1), SIMP chHES-3 cells (Lane 2), and COMP chHES-3 cells (Lane 3).



karyotype 47,XX1X,dup(1)(p32p36),t(1;6;4) (q25;

q23;p16), ins(4;1)(p16;q21q25),t(7;8)(q22;q22) (Fig.

2G); from P188 to P197, the karyotype further

changed in which a novel marker chromosome and

disomy X appeared but the trisomy X disappeared

(Fig. 2H). FISH analysis (Figs. 3A–3D) further con-

firmed some of the changes in the karyotypes. Fur-

thermore, CGH analysis of chHES-3 cells at P96

and at P187 both showed only gain of 1p32-p36

(Figs. 4A and 4B) although routine G-banding

revealed more complex chromosome rearrange-

ment. All those results suggest that chromosomal

abnormalities in hESCs can be passed on to the

next generation and accumulate.

To clarify whether high-density feeder cells

have a detrimental effect on karyotypic stability,

feeder cells at different density were used to cul-

ture chHES-3 cells. When karyotypically normal

chHES-3 cells of early passages (P19) were thawed

and cultured on lower density feeder layer of 2.5 3104 cells/cm2, they maintained a normal karyotype

up to P74; when cultured on feeder cells at higher

density of 4.0–5.0 3 104 cells/cm2, they also

acquired the aberrant karyotype with the simple

duplication in chromosome 1 (triplet experiments),

which when were transferred back onto lower den-

sity feeder layer, only the duplication karyotype

were observed without further changes up to P114.

Figure 2. G-banding of chHES-3 cells at different passages. (A) Nor-mal karyotype of 46,XX at passage 27. (B) Abnormal karyotype of46,XX,dup (1)(p32p36) at passage 34. (C) Abnormal karyotype withcomplex chromosome rearrangement involving chromosomes 1, 4, and6, and a reciprocal translocation between chromosomes 7 and 8 wasfound at passage 44–99, seen as 46,XX, dup(1)(p32p36),t(1;6;4)(q25;q23;p16),ins(4;1)(p16;q21q25), t(7;8)(q32;q13), and was also foundin 2 of 19 metaphases at P114. (D) One of 19 metaphases at P114 was46,XX,dup(1)(p32p36),t(1;6;4)(q25;q23;p16),ins(4;1)(p16;q21q25),

t(7;8)(q32;q13),1mar(2?). (E) One of 19 metaphases at passage 114was 46,XX,dup(1)(p32p36),t(1;6;4)(q25;q23;p16),ins(4;1)(p16;q21q25),t(7;8)(q32;q13),1mar(2?),1mar(6?),1mar(12?). (F) Eleven of 19 meta-phases at passage 114 were 47,XX1X,dup(1)(p32p36),t(1;6;4)(q25;q23;p16),ins(4;1)(p16;q21q25),t(7;8)(q22;q22). (G) There existedonly one karyotype described in (f) at passages 142–153. (H) Thekaryotype at passages 188–197 became 46,XX,dup(1)(p32p36),t(1;6;4)(q25;q23;p16),ins(4;1)(p16;q21q25),t(7;8)(q22;22),1mar(15?).


670 YANG ETAL.

All the above mentioned results suggest that

higher-density feeder layers might have an effect

on chromosomal changes.

Lower Differentiation of Karyotypically Aberrant

chHES-3 Cells In Vivo

Previous reports have shown that karyotypic

changes give rise to a malignant phenotype of

hESCs in vivo (Andrews et al., 2005). To deter-

mine the tumorigenicity of aberrant chHES-3, we

compared the teratomas derived from chHES-3

cells with a complex karyotype at P96 (COMP)

with those derived from chHES-3 cells with a nor-

mal karyotype at P17. In xenografts from COMP

chHES-3 cells, there were tissues from three germ

layers, predominantly immature mesenchymal tis-

sues, and primitive neural tissues with neural tube-

like rosette structure (Fig. 5A) in which there were

more cellular layers with apoptosis bodies and nu-

clear division metaphases (Fig. 5B). In every sec-

tion there were more than three low power fields

(403) containing the primitive neural tissues.

According to the diagnostic criteria defined by the

World Health Organization (Mikuz, 2002), the

tumors were classified as immature teratoma of

grade III. And in the teratoma, nested regions with

Oct4-positive cells were also observed (Fig. 5C).

Compared to the teratoma derived from the

COMP chHES-3 cells, the teratoma derived from

the normal chHES-3 presented higher degree of

differentiation with few immature neural tissues

(Fig. 5D) and less cellular layers in the immature

neural tissue (Fig. 5E), as well as Oct4-negative

(Fig. 5F). Similar to in vivo, the abnormal chHES-3

Figure 3. FISH analysis of chHES-3 cells at different passage. FISHanalysis at passage 34 using WCP1 (red ) and WCP4 (green) showed anadditive red signal on one chromosome 1 (red), while the other chro-mosome 1 is normal, confirming that the additive fragment originatedfrom chromosome 1 (i). FISH analysis at passage 65 using WCP1 (red)and WCP6 (green) (j), WCP1 (red) and WCP4 (green) (k), and WCP7(red) and WCP8 (green) (l), respectively. In (j), there is a red signal(Chr 1) on der(6) ( green), and a red signal (Chr 1) and a green signal

(Chr 6) on der(4); In (k), there is a green signal (Chr 4) on der(1) (red),a red signal (Chr 1) on der (6), and a green signal (Chr 6) and a red sig-nal (Chr 1) on der(4); In (l), there is a red signal (Chr 7) on der(8)(green) and a green signal (Chr 8) on der(7) (red). These results con-firm that there are complex rearrangements involving chromosomes 1,4, and 6, an insertional between chromosome 1 and 4, and a reciprocaltranslocation between chromosome 7 and 8.



cells make more rosettes in vitro (data not shown).

Furthermore, the aberrant chHES-3 cells formed

embryoid bodies in vitro and had capacity of differ-

entiation to three germ layers after 3–4 weeks of

differentiation (Fig. 6).

Higher Ratio of S Phase and Lower Ratio of

Apoptosis in Karyotypically Aberrant chHES-3 Cells

We systematically compared normal chHES-3,

SIMP chHES-3, and COMP chHES-3 cells in cell

cycle distribution and cell apoptosis ratios. The

chHES-3 cells were detected by ES markers

including the SSEA-4 before analysis. In the popu-

lation of ES cells, 84–96% cells are SSEA-4 posi-

tive. The percentage of cells in S-phase gradually

increased from 33.1% 6 1.27% (normal chHES-3

cells) to 58.2% 6 0.71% (SIMP chHES-3 cells) and

to 72.7% 6 1.5% (COMP chHES-3cells Fig. 7A),

while the apoptosis ratio sharply decreased from

10.3% 6 0.4% (normal chHES-3 cells) to 4.0% 60.9% (SIMP chHES-3 cells) and to 1.0% 6 0.3%

(COMP chHES-3 cells Fig. 7B). To detect the

capacity of apoptosis resistance, normal chHES-3,

SIMP chHES-3, and COMP chHES-3 cells were

Figure 4. CGH analysis of COMP chHES-3 cells at passage 96 (m) and P187 (n). An obvious gain (green)of 1p32?p36 was found.


672 YANG ETAL.

cultured in 0.01 lm mitomycin C, the apoptosis

ratios were 13.3% 6 0.89%, 4.83% 6 0.93%, and

1.06% 6 0.12% respectively. Furthermore, during

routine culture, the ratios of single clone-formation

were 0.165% 6 0.01%, 0.18% 6 0.03%, and 0.96%

6 0.11%, respectively. All these results indicate

that compared to that of normal chHES-3 cells, the

aberrant chHES-3 cells exhibited gradually

increased capacities of proliferation and antiapop-

tosis with increased complexity of karyotype.

Gene Expression Profile Analysis

The SIMP (P53) and COMP (P182) chHES-3

cells were selected as karyotypically aberrant sam-

ples to detect gene expression profiles, with the

normal chHES-3 (p30) cells being used as control.

A simple present/absent analysis of gene expres-

sion in the samples is shown as Venn diagrams

(Fig. 8). In total, 12,045, 11,001, and 11,375 genes

were expressed in the normal chHES-3 cells, SIMP

cells, and COMP cells, respectively, showing that

the numbers of expressed genes were clearly dif-

ferent among them. More genes were expressed in

the normal chHES-3 cells when compared with

those in two aberrant chHES-3 samples. Of them,

965, 310, and 346 genes were uniquely expressed

in normal chHES-3 cells, SIMP chHES-3 cells, and

COMP chHES-3 cells, respectively.

Upregulation of Pluripotency-Related Genes and

Downregulation of Differentiation-Related Genes

The expression of all 10 genes related to pluripo-

tency and self-renewal mentioned in a previous

study (Mitalipova et al., 2005), including POU5F1,

Figure 5. Histopathological and immunohistological analyses oftumors. (A–C) Teratoma derived from COMP chHES-3 cells, showing alarge area of primitive nerve tissue containing many neural tube-likestructre (arrows) (A), neural tube-like structure with multilayer of cells(B) and nested region of Oct4-positive cells (C). (D–F) Teratoma

derived from normal chHES-3 cells, showing well differentiated tissues(D), neural tube-like structure with only a few layers of cells (E) and noOct4-positive cells (F). Insert shows the DAPI staining for cell nuclei.(triplet experiments).

Figure 6. Differentiation analysis of hESC-derived embryoid bodies in vivo. (A) Embryoid body derivedfrom aberrant karyotypic HES cells. (B) Expression of genes related to the three germ layers in differentcells from embryonic body. [Color figure can be viewed in the online issue, which is available at www.interscience.wiley.com.]



Nanog, LDB2, GABRB3, FGF4, LEFTY2, LDB2,

DNMT3B, BUB1, and CD9, were gradually upre-

gulated with increased complexity of chromosomal

aberration, while the expression of differentiation-

related genes such as FN1, MSI2, NEDD1, SEPT2,

NEFH, NEDD4, PAX6, OTX2, MCFD2, and

SOX3 in 21 ectoderm development-related genes,

EOMES, FLT1, HLA-C, MYL4, PECAM1, T,

THBS1, THBS2, and VCAM1 in 28 mesoderm de-

velopment-related genes, and AFP, CER1, FOXA2,

GATA4, GATA6, and HNF4A in 7 endoderm devel-

opment-related genes (Mitalipova et al., 2005) had a

tendency to decrease with increased karyotypic

complexity, whereas the expression level was

unchanged for the rest of them (Table.3). Hierarchi-

cal cluster analysis of pluripotency-related genes

showed that normal chHES-3 cells and SIMP

chHES-3 cells are more closely related to each other

than to the COMP chHES-3 cells (Fig. 9A).

Downregulation of Tumor-Suppression Genes and

Upregulation of Oncogenes

Thirteen tumor suppressor genes, including

PTPRG, PTCH, SMAD4, PTEN, RERE,

RPL10A, TIMP1, CDH1, APC, TP53, BRCA1,

MSH2, and NME1, were expressed in all three

samples. Of these, SMAD4, RERE, CDH1, APC,

TP53, and BRCA1 exhibited a gradually decreased

expression with increased karyotypic complexity.

Hierarchical cluster analysis demonstrated that

normal and SIMP cells were closely related to each

other but distinct from COMP cells (Fig. 9B).

Analysis of 10 oncogenes, including FGFR1,

MDM2, BCL2, LMO2, ERBB2, TPM3, NTRK1,

MET, CDK4, and LMO1, in SIMP, COMP chHES-3

cells, and normal chHES-3 cells showed no significant

difference in their expression levels (data not shown).

However, the karyotype and CGH analysis showed

that duplication of 1p32–1p36 was the only dosage

variation of the genome in SIMP chHES-3 cells,

which might be an important event in malignant

transformation and genomic instability. Therefore,

we further investigated the expression of oncogenes

in this region, including HKR3, DJ1, FGR, LCK,

MPL, BLYM, MYCL1, BLF, TAL1, and JUN by

using semiquantitative PCR and real-time PCR. Of

these, the expression levels of HKR3 and LCK in

SIMP chHES-3 cells were higher than in the normal

chHES-3 cells, while the expression levels of the

other genes showed no difference (Fig. 10).

Activation of Wnt Signaling Pathway

Previous studies have shown that important

genes related to the Wnt signaling pathway are

involved in self-renewal of hESCs; and tumorigen-

esis exhibit significantly differential expression in

stem cells, cancer stem cells, and cancer (Reya

et al., 2001; Ying et al., 2003; Sato et al., 2004). In

this study, the microarray results showed that of all

analyzed signaling pathways, the intracellular Wnt

Figure 7. Analyses of cell cycle distribution and cellular apoptosis.(A) The percentage of S-phase cells of normal chHES-3 cells, simpleduplication (SIMP) chHES-3 cells, and complex karyotype (COMP)chHES-3 cells (three independent experiments) (mean 6 s.d., n 5 3; *P< 0.05; **P < 0.05 by ANOVA). (B) The percentage of apoptotic cellsin normal, SIMP and COMP chHES-3 cells (mean 6 s.d., n 5 3; *P <0.05; **P < 0.05 by ANOVA).

Figure 8. The number of expressed genes in three types of cells.Normal chHES-3 cells, simple duplication chHES-3 cells, and complexkaryotypic chHES-3 cells are denoted by Normal, SIMP, and COMP,respectively. [Color figure can be viewed in the online issue, which isavailable at www. interscience.wiley.com.]


674 YANG ETAL.

signaling pathway was preferentially activated with

increased karyotypic complexity. Of all differen-

tially expressed genes related to the Wnt signaling

pathway, the inhibitor GSK3b and Wnt-antagonists

SFRP1, SFRP2, and FRZB were gradually down-

regulated with increased karyotypic complexity,

while b-catenin and its downstream oncogenes Jun

and cyclin D1, and other genes, such as E-cadherin

and the Wnt receptor Fzd2, were gradually upregu-

lated (Table 3). However, the Wnt genes were not

expressed differentially. These findings were fur-

ther supported by real-time PCR analysis (Figs.

11A–11F). These results suggested that activated

Wnt signaling pathway might be a main reason for

malignant transformation of chHES-3 cells, similar

to the results observed in other tumorigenesis in

tumors (Gat et al., 1998; Moon et al., 2002).

Figure 9. Hierarchical cluster analyses. (A) pluripotency-related genes and (B) tumor-suppressiongenes. The normal chHES-20 cells (normal) and the simple duplication (SIMP) chHES-3 cells are moreclosely related to each other than to the complex karyotype (COMP) chHES-3 cells.

Figure 10. Real-time RT-PCR analysis of LCK and HKR3 demon-strated a 2.4-(LCK) and a 1.3-(HKR3) fold expression increase in thesimple duplication (SIMP) chHES-3 cells compared with the normalchHES-3 cells, respectively (n 5 3).



Verification of Microarray Results by Real-Time

RT-PCR

To verify the results of the gene chips, we con-

firmed the expression changes of some important

genes by real time RT-PCR with primers listed in

Table 2. The results are shown in Figure 12 and

summarized in Table 4.

DICUSSION

More and more evidence has shown that the

karyotype of hES cells change during long-term

culture. The typical changes are additional copies

of chromosomes 12, 17, 1, and X, similar to germ

cell tumors, such as malignant EC (Sandberg et al.,

1999; Andrews, 2002;). In this study, the karyotypic

changes started with a duplication of 1p32-36. Fur-

thermore, this study is the first detailed report on

the karyotypic change process, exhibiting progres-

sive chromosomal aberrations during long-term

culture; from an early simple duplication to a late

complex karyotypes similar to those in tumors

through gradual accumulation of new abnormal-

ities. Draper (2004) indicated that karyotypic

changes of chromosomes 12, 17, and X confer hES

cells with a tumorigenic capacity (Baker et al.,

2006). Our results show that karyotypic changes of

chromosome 1 might be related to tumorigenesis,

not least because there are many oncogenes and

tumor suppression genes on chromosome 1. Karyo-

typic change affected the function of ES cells. At

the early stage of karyotypic changes, LCK,

HKR3, two oncogenes mapping to the duplication

region of 1p32-36, were upregulated. Furthermore,

genes important for proliferation in the canonical

Wnt pathway (Wnt4) and survival pathways

(EGF), or cell apoptosis (FAF1, TP73, TNFRSF8,

TNFRSF9, TNFRSF14, CASP9, and JUN), or

cell cycle regulation (CDKN2C, CDC42,

CDC2L1, DDEFL1) map to 1p32-36. This might

be one reason for chromosomal duplication and the

high cell proliferation.

Previous studies indicated that the karyotypic

changes in the hESCs during long-term culture in

vitro were adaptive changes to the suboptimal cul-

ture and the adaptive karyotypic changes conferred

hESCs with more power of self-renewal (Buzzard

et al., 2004; Baker et al., 2006). In present studies,

the aberrant karyotypic chHES-3 cells exhibited

the characteristics similar to normal ES cells, such

as the special cellular markers, alkaline phospha-

tase activity, and telomerase activity, whereas the

results from the apoptosis analysis, distribution of

cell cycle, and in vitro differentiation demonstrated

TABLE 3. The Expression Changes of Wnt SignallingPathway-Related Genes in Microarrary

Gene title

Changea

SIMPb COMPc

Glycogen synthasekinase 3b

NCd D

Catenin (cadherin-associated protein),b 1, 88 kDa

De I

Casein kinase 1, epsilon If IWingless-type MMTVintegration sitefamily, member 6

NC NC

Frizzled homolog 6 D NCWingless-type MMTVintegration site family,member 1

D D

Wingless-type MMTVintegration sitefamily, member 4

D D

Frizzled homolog 3 (Drosophila) D DFrizzled homolog 7 (Drosophila) D DFrizzled homolog 2 (Drosophila) I IFrizzled homolog 6 (Drosophila) D DFrizzled homolog 1 (Drosophila) NC DFrizzled homolog 5 (Drosophila) NC DFrizzled homolog 2 (Drosophila) D DFrizzled homolog 3 (Drosophila) D DFrizzled homolog 10 (Drosophila) D DFrizzled homolog 5 (Drosophila) NC DCyclin D2 D DProtein kinase C, iota NC NCPlatelet-activating factoracetylhydrolase, isoform Ib,a subunit 45 kDa

D D

Platelet-activating factoracetylhydrolase,isoform Ib,b subunit 30 kDa

D D

Protein kinase D1 NC DLow density lipoproteinreceptor-related protein 6

NC I

Ras homolog gene family,member A

NC D

Adenomatosis polyposis coli NC NCras-related C3 botulinumtoxin substrate 1

I NC

v-jun sarcoma virus 17oncogene homolog (avian)

I I

Secreted frizzled-relatedprotein 1

D D

Secreted frizzled-relatedprotein 2

D D

Frizzled-related protein D DJUN I I

aChange, the change tendency of abnormal hESCs compared to normal

hESCs.bSIMP, chHES3 cells with a simple duplication karyotype.cCOMP, chHES3 cells with a complex karyotype.dD, decrease.eI, increase.fNC, no change.


676 YANG ETAL.

that their power of antiapoptosis and proliferation

gradually increased with the increased karyotypic

complexity. In a previous study, karyotypic changes

in mouse ES cells correlated with a reduced ability

to colonize the germ line in the chimeric mice,

implying that karyotypic changes decreased the

capacity of cellular differentiation (Longo et al.,

1997). Recent studies have also shown that adapt-

ive karyotypic changes confer hESCs with more

power of self-renewal at expense of cellular death

and differentiation capacity (Keith, 2004). In our

experiments of in vivo differentiation, the imma-

ture neural tissues dominantly existed in the tera-

toma derived from the COMP chHES-3 cells and

were more than that derived from karyotypically

normal chHES-3 cells, indicating that the karyo-

typic changes might lead to the restricted differen-

tiation power to the neural tissues, and further

illustrating that the karyotypically aberrant

chHES-3 cells had stronger power of self-renewal

and proliferation at expense of cellular death and

differentiation capacity. Consistent with this, the

results of gene chips and real-time RT-PCR indi-

cated that the expression of genes related to self-

renewal, such as POU5F1, Nanog, LDB2, GABRB3,

FGF4, LEFTY2, LDB2, DNMT3B, BUB1, and

CD9, increased in karyotypically aberrant chHES-3

cells, wheras the expression of genes related to dif-

ferentiation of the three germ layers decreased.

From the tissue characteristics in teratoma derived

from the karyotypically aberrant chHES-3 cells we

propose that the chHES-3 cells are gradually trans-

formed to malignant cells due to the dysregulated

stem cell self-renewal and proliferation/survival,

with the accumulated karyotypic changes.

More evidence shows that the Wnt signaling

pathway, which normally controls stem cell fate, is

activated in tumorigenesis (McWhirter et al., 1999;

Figure 11. Relative expression level of genes in Wnt signal pathwayin karyotypically aberrant HES cells versus normal HES cells. (A) mRNAlevels of FED2, (B) E-cadherin, (C) FRZB, (D) SFRP1, and (E) SFRP2were detected by real-time RT-PCR of normal chHES-3 cells (n 5 3),

SIMP chHES-3 cells (n 5 3), and COMP chHES-3 cells (n 5 3). The rela-tive expression levels were determined by normalizing the DCt valuesof aberrant HES cells against the average DCt values of normal HES cellsfor the specified genes.



Derksen et al., 2004). In this study, the results of

the microarray and real-time RT-PCR analysis

demonstrate that in all investigated signaling path-

ways the intracellular Wnt signaling pathway was

preferentially activated with increased karyotypic

complexity, while no differentiated expression of

Wnt genes was detected in karyotypically normal

and aberrant hESCs. We conclude that the Wnt

signaling pathway might be activated by Wnt pro-

teins from the feeder cells because our previous

proteomics results on feeder cells showed that the

Kunming-White mouse feeder cells produced

Wnt-3 protein (Xie et al., 2004). We further investi-

gated the expression of Wnt family genes in the

feeder cells and found that 13 of 19 Wnt family

genes were expressed, including WNT1 etc (data

not shown) and that the expression amount of most

expressed genes were positively correlated to the

density of feeder cells, including Wnt2 etc (data

not shown). We propose that excessive production

of Wnt proteins from high-density feeder cells

might be one of the main reasons for cellular tumor

progression in chHES-3 cells. Whether Wnt pro-

teins or other factors are involved in this process

remaine to be determined.

In conclusion, chHES-3 cells undergo deregula-

tion of self-renewal and dysfunction of related

genes in long-term culture adaptation, leading to

malignant transformation; the activation of

Wnt signaling pathway might be involved in this

process.

ACKNOWLEDGMENTS

The authors would like to thank Dr. Alan Troun-

son for helpful suggestions and kindly revising

manuscript, Dr. Hsiao Chang Chan for helpful

comments and suggestions and Dr. Liang Hu for

kindly carrying out the CGH analysis.

REFERENCES

Andrews PW. 2002. From teratocarcinomas to embryonic stem cells.Philos Trans R Soc Lond B Biol Sci 357:405–417.

Andrews PW. 2006. The selfish stem cell. Nat Biotechnol 24:325–326.

Andrews PW, Matin MM, Bahrami AR, Damjanov I, Gokhale P,Draper JS. 2005. Embryonic stem cells and embryonal carcinoma(EC) cells: Opposite sides of the same coin. Biochem Soc Trans33:1526–1530.

Babaie Y, Herwig R, Greber B, Brink TC, Wruck W, Groth D, Leh-rach H, Burdon T, Adjaye J. 2007. Analysis of Oct4-dependenttranscriptional networks regulating self-renewal and pluripotencyin human embryonic stem cells. Stem Cells 25:500–510.

Baker DE, Harrison NJ, Maltby E, Smith K, Moore HD, Shaw PJ,Heath PR, Holden H, Andrews PW. 2006. Adaption to culture ofhuman embryonic stem cells and oncogenesis in vivo. Nat Bio-technol 25:207–215.

Becker KA, Ghule PN, Therrien JA, Lian JB, Stein JL, van WijnenAJ, Stein G.S. 2006. Self-renewal of human embryonic stem cellsis supported by a shortened G1 cell cycle phase. J Cell Physiol209:883–893.

Buzzard JJ, Gough NM, Crook JM, Colman A. 2004. Karyotype ofhuman ES cells during extended culture. Nat Biotechnol 22:381–382.

Cowan CA, Klimanskaya I, McMahon J, Atienza J, Witmyer J,Zucker JP, Wang S, Morton CC, McMahon AP, Powers D, Melton

Figure 12. Confirmation of gene expression changes of selectedgenes by real-time RT-PCR. Ratios are represented as log-ratio (base 2),with values above zero denoting upregulation of gene expression andbelow zero denoting downregulation (n 5 3).

TABLE 4. The DCt Values of Genes Obtained fromthe Real-Time RT-PCR

Genes

Normal ESa SIMPb COMPc

Mean 6SD Mean 6SD Mean 6SD

Pluripotency-related genenanog 9.70 1.44 9.51 0.54 9.03 2.74POU5F1 5.32 0.90 5.07 0.85 4.88 2.39Sox2 9.60 1.95 9.68 1.98 9.50 0.72

Chromosomal instability related genesBUB1 16.20 0.79 14.05 0.68 11.88 2.32TERF1 9.82 0.30 9.35 1.04 9.05 0.13

Wnt signalling pathway-related genesGSK-3Beta 12.77 0.16 14.27 1.30 14.67 0.78Beta-catenin 25.05 1.08 22.38 0.91 19.77 0.46Cyclin D1 11.77 1.22 10.73 0.25 9.88 0.71

OncogenesFGFR1 11.95 0.53 10.16 0.33 11.18 0.68CDK4 11.22 1.18 10.10 0.79 8.43 0.24V-jun 16.72 0.32 14.71 0.64 12.95 0.17

Tumor-suppression genesTp53 10.97 1.07 11.08 0.24 11.31 0.20NF1 20.23 0.85 19.80 0.32 19.98 0.80

Three germ layer-related genesAFP 17.75 0.79 18.35 0.43 18.64 0.44PAX6 10.27 0.31 13.37 0.94 17.67 0.80EOMES 13.67 0.06 16.18 0.32 20.00 2.79

aNormal ES, chHES20 cells with a normal karyotype.bSIMP, chHES3 cells with a simple duplication karyotype.cCOMP, chHES3 cells with a complex karyotype.


678 YANG ETAL.

DA. 2004. Derivation of embryonic stem-cell lines from humanblastocysts. N Engl J Med 350:1353–1356.

Derksen PW, Tjin E, Meijer HP, Klok MD, MacGillavry HD, vanOers MH, Lokhorst HM, Bloem AC, Clevers H, Nusse R, vander Neut R, Spaargaren M, Pals ST. 2004. Illegitimate WNT sig-naling promotes proliferation of multiple myeloma cells. ProcNatl Acad Sci USA 101:6122–6127.

Draper JS, Smith K, Gokhale P, Moore HD, Maltby E, Johnson J,Meisner L, Zwaka TP, Thomson JA, Andrews PW. 2004. Recur-rent gain of chromosomes 17q and 12 in cultured human embry-onic stem cell. Nat Biotechnol 22:53–54.

Enver T, Soneji S, Joshi C, Brown J, Iborra F, Orntoft T, ThykjaerT, Maltby E, Smith K, Dawud RA. 2005. Cellular differentiationhierarchies in normal and culture-adapted human embryonicstem cells. Hum Mol Genet 14:3129–3140.

Gat U, DasGupta R, Degenstein L, Fuchs E. 1998. De Novo hairfollicle morphogenesis and hair tumors in mice expressing a trun-cated b-catenin in skin. Cell 95:5–14.

Hoffman LM, Carpenter MK. 2005. Characterization and culture ofhuman embryonic stem cells. Nat Biotechnol 23:699–708.

Imreh MP, Gertow K, Cedervall J, Unger C, Holmberg K, Szoke K,Csoregh L, Fried G, Dilber S, Blennow E. 2006. In vitro cultureconditions favoring selection of chromosomal abnormalities inhuman ES cells. J Cell Biochem 99:508–516.

Inzunza J, Sahlen S, Holmberg K, Stromberg AM, Teerijoki H,Blennow E, Hovatta O, Malmgren H. 2004. Comparative genomichybridization and karyotyping of human embryonic stem cellsreveals the occurrence of an isodicentric X chromosome afterlong-term cultivation. Mol Hum Reprod 10:461–466.

Keith WN. 2004. From stem cells to cancer: balancing immortalityand neoplasia. Oncogene 23:5092–5094.

Kim NW, Piatyszek MA, Prowse KR, Harley CB, West MD, Ho PL,Coviello GM, Wright WE, Weinrich SL, Shay JW. 1994. Specificassociation of human telomerase activity with immortal cells andcancer. Science 266:2011–2015.

Klein CA, Schmidt-Kittler O. 1999. Comparative genomic hybrid-ization, loss of heterozygosity, and DNA sequence analysis of sin-gle cells. Proc Natl Acad Sci USA 96:4497–4452.

Livak SN, Schmittgen TD. 2001. Analysis of relative gene expres-sion data using real-time quantitative PCR and the 2-DDct meth-ods. Methods 25:402–403.

Longo L, Bygrave A, Grosveld FG, Pandolfi PP. 1997. The chromo-some make-up of mouse embryonic stem cells is predictive of so-matic and germ cell chimaerism. Transgenic Res 6:321–328.

McWhirter JR, Neuteboom ST, Wancewicz EV, Monia BP, DowningJR, Murre C. 1999. Oncogenic homeodomain transcription factorE2A-Pbx1 activates a novel WNT gene in pre-B acute lympho-blastoid leukemia. Proc Natl Acad Sci USA 96:11464–11469.

Mikuz G. WHO classification of testicular tumors. 2002. Verh DtschGes Pathol 86:67–75.

Mitalipova MM, Rao RR, Hoyer DM, Johnson JA, Meisner LF,Jones KL, Dalton S, Stice SL. 2005. Preserving the genetic integ-rity of human embryonic stem cells. Nat Biotechnol 23:19–20.

Moon RT, Bowerman B, Boutros M, Perrimon N. 2002. The prom-ise and perils of Wnt signaling through b-catenin. Science296:1644–1646.

Reya T, Morrison SJ, Clarke MF, Weissman IL. 2001. Stem cells,cancer, and cancer stem cells. Nature 414:105–111.

Sandberg AA, Meloni AM, Suijkerbuijk RF. 1999. Reviews of chro-mosome studies in urological tumors. III. Cytogenetics and genesin testicular tumors. J Urol 155:1531–1556.

Sato N, Meijer L, Skaltsounis L, Greengard P, Brivanlou AH. 2004.Maintenance of pluripotency in human and mouse embryonicstem cells through activation of Wnt signaling by a pharmacologi-cal GSK-3-specific inhibitor. Nat Med 10:55–63.

Thomson JA, Itskovitz-Eldor J, Shapiro SS, Waknitz MA, SwiergielJJ, Marshall VS, Jones JM. 1998. Embryonic stem cell linesderived from human blastocysts. Science 282:1145–1147.

Weinrich SL, Pruzan R, Ma L, Ouellette M, Tesmer VM, Holt SE,Bodnar AG, Lichtsteiner S, Kim NW, Trager JB, Taylor RD, Car-los R, Andrews WH, Wright WE, Shay JW, Harley CB, Morin GB.1997. Reconstitution of human telomerase with the templateRNA component hTR and the catalytic protein subunit hTRT.Nat Genet 17:498–502.

Xie CQ, Lin G, Luo KL, Luo SW, Lu GX. 2004. Newly expressedproteins of mouse embryonic fibroblast irradiated to be inactive.Biochem Biophys Res Commun 315:581–588.

Ying QL, Nichols J, Chambers I, Smith A. 2003. BMP induction ofId proteins suppresses differentiation and sustains embryonicstem cell self-renewal in collaboration with STAT3. Cell 115:281–292.

Zeng X, Chen J, Liu Y, Luo Y, Schulz TC, Robins AJ, Rao MS,Freed WJ. 2004. BGO1V a variant human embryonic stem cellline which exhibits rapid growth after passaging and reliable do-paminergic differnetiation. Restor Neurol Neurosci 22:421–424.



Documents

Tumor progression of culture-adapted human embryonic stem cells during long-term culture