Human embryonic stem cells (HESCs) have two
unique properties: self-renewal, the ability to pro-
liferate indefinitely while maintaining their cellular
identity; and pluripotency, the ability to differentiate
into all of the cell types that comprise the embryo proper.
These traits make HESCs promising for future regenera-
tive medicine, but the same traits also make them tum-
origenic, and consequently hinder the fulfilment of their
clinical potential. Thus, HESCs could be aptly described
as ‘double-edged swords’, as their defining characteristics
make them both powerful and dangerous.
The tumorigenicity of HESCs
HESCs share cellular and molecular phenotypes with
tumour cells and cancer cell lines1–3. Among these are
rapid proliferation rate4, lack of contact inhibition5, a pro-
pensity for genomic instability6,7, as well as high activity of
telomerase8, high expression of oncogenes such as MYC9
and KLF4 (REF. 10), and remarkable similarities in their
overall gene expression patterns11–13, microRNA (miRNA)
signatures14 and epigenetic status15. When injected
into immunodeficient mice, HESCs form teratomas5.
These tumours are so characteristic of HESCs that they
have become the most stringent test for pluripotency in
human cells. Indeed, treatment attempts with embryonic
stem cells in animal models were shown to be fatal owing
to the formation of teratoma-like tumours16.
The above-mentioned features apply to normal dip-
loid HESCs in their untransformed state; however, HESCs
can also undergo transformation in culture. Several
studies have demonstrated that culture-adapted HESCs
can form more aggressive tumours, which might be clas-
sified as teratocarcinomas, the malignant counterparts of
However, the risk of formation of teratocarcinomas
on transplantation of HESC-derived cells is not limited
to transformed aneuploid HESCs. Studies have shown
that mouse embryonic stem cells (MESCs) with a nor-
mal karyotype form teratocarcinomas when injected into
immunodeficient mice21. Thus, it is possible that trans-
planting HESCs into humans could result in malignant
teratocarcinomas, rather than in benign teratomas. When
injected into engrafted human fetal tissues in severe com-
bined immunodeficient (SCID) mice, HESCs indeed
generated primitive, undifferentiated tumours22. Given
the obvious inability to directly test this in humans, this
concern remains largely unsolved. However, regardless
of the risk of developing malignant tumours, the forma-
tion of benign teratomas on transplantation of HESCs
or HESC-derived cells into humans would also be highly
alarming and unacceptable. Therefore, the tumorigenic-
ity of HESCs is a major hurdle, which must be confronted
before achievements of this field of research can be safely
translated into the clinic. For recent reviews on the
tumorigenicity of HESCs see REFS 2,3,20.
HiPSCs: a new source for pluripotent cells
The discovery of human induced pluripotent stem cells
(HiPSCs)23–25 has revolutionized the field of pluripotent
stem cell research for two main reasons: it changed the
Stem Cell Unit, Department
of Genetics, Silberman
Institute of Life Sciences,
The Hebrew University,
Jerusalem 91904, Israel.
Correspondence to N.B.
10 March 2011
Benign tumour that is
composed of differentiated
tissues from all three germ
Tumour composed of a mixture
of differentiated tissues of the
three germ layers. Contains foci
of completely undifferentiated
cells, called embryonal
carcinoma cells, and is highly
The tumorigenicity of human
embryonic and induced pluripotent
Uri Ben-David and Nissim Benvenisty
Abstract | The unique abilities of human pluripotent stem cells to self-renew and to
differentiate into cells of the three germ layers make them an invaluable tool for the future of
regenerative medicine. However, the same properties also make them tumorigenic, and
therefore hinder their clinical application. Hence, the tumorigenicity of human embryonic
stem cells (HESCs) has been extensively studied. Until recently, it was assumed that human
induced pluripotent stem cells (HiPSCs) would behave like their embryonic counterparts in
respect to their tumorigenicity. However, a rapidly accumulating body of evidence suggests
that there are important genetic and epigenetic differences between these two cell types,
which seem to influence their tumorigenicity.
268 | APRIL 2011 | VOLUME 11
© 2011 Macmillan Publishers Limited. All rights reserved
The expression of specific
genes from either the maternal
or the paternal allele.
CpG island shore
DNA sequence that flanks CpG
perception of cellular reprogramming, showing that the
plasticity of somatic cells is much greater than had previ-
ously been thought; and it offered an appealing solution
to the likely immune rejection of HESC-derived cells on
their transplantation into an unmatched patient, thus
providing new and exciting avenues for patient-specific
cell therapy. Importantly, HiPSCs also provide a possible
solution to the ethical objections that have been raised
against the use of HESCs, which is a highly controversial
topic in many countries.
Although constituting a huge leap towards overcoming
immunogenic and ethical obstacles, the translation of
HiPSCs into the clinic faces the same substantial tumori-
genicity problem as that of HESCs. Sharing with HESCs
their basic properties of self-renewal and pluripotency,
HiPSCs are doomed to share with them the other ‘edge
of the sword’. Indeed, HiPSCs exhibit the cellular and
molecular phenotypes that make HESCs resemble cancer
cells (discussed above), and form benign teratomas on
injection into immunodeficient mice. However, HESCs
and HiPSCs (as well as their mouse counterparts) are
not identical, and rapidly accumulating evidence sug-
gests that there are considerable differences between
these two pluripotent cell types, including important
aspects such as global gene expression26, epigenetic land-
scape27–29 and genomic imprinting30. In fact, HiPSCs are
relevant for research into the tumorigenicity of pluripo-
tent cells, as the process of cellular reprogramming of
somatic cells into pluripotent cells can teach us much
about the tumorigenicity of human pluripotent cells in
general. As these cells are the newest and most promising
type of pluripotent cells, it is crucial to specifically study
the tumorigenicity of HiPSCs and to compare it to the
tumorigenicity of HESCs, which are the ‘gold standard’
for human pluripotent cells.
Reprogramming genes and HiPSC tumorigenicity
HiPSCs were first derived by the transduction of
fibroblasts with integrating viruses carrying four tran-
scription factors: OCT4, SOX2, MYC and krupple-like
factor 4 (KLF4)24. Although MYC is a well-established
oncogene31,32, the other three transcription factors are
also known to be highly expressed in various types of
cancer33–40. Indeed, one study found significant overex-
pression of at least one of these reprogramming genes
in 18 of the 40 cancer types investigated41. Furthermore, in
specific types of tumours, these genes were found to be
associated with tumour progression and a poor progno-
sis41. Reactivation of the reprogramming factors has also
been shown to predispose iPSCs to genomic instability42.
Recently, new methods have been developed to repro-
gramme human somatic cells without MYC25,43 or with a
transformation-deficient MYC44, and by combining only
some of the reprogramming transcription factors with
chemical inhibitors45–48. However, the fact remains that
OCT4, SOX2, MYC and KLF4 reside at the heart of the
reprogramming process, stressing the potential risks of
these new pluripotent cells.
Further evidence for the similarity between the repro-
gramming of somatic cells to iPSCs and tumorigenesis
was provided in a series of papers that demonstrated
how downregulation of tumour suppressors in the p53
pathway increases the efficiency of the reprogramming
process and enables reprogramming with only two fac-
tors (OCT4 and SOX2)49–53 (reviewed in REF. 54). A recent
report found that, in respect to their overall miRNA
signature, the expression levels of miRNAs that belong
to the p53 network make some HiPSC lines (and par-
tially reprogrammed cells) more similar to cancer cell
lines14. In addition, as telomerase activity is essential for
efficient reprogramming, HiPSCs develop longer telom-
eres and acquire the epigenetic marks of undifferentiated
cells55,56, similar to the transformation of cancer cells57.
Importantly, however, cellular reprogramming is not
equivalent to cancerous transformation. A recent whole
genome-wide study of tissue- and cancer-specific meth-
ylation of CpG island shores showed an inverse correlation
between HiPSCs and cancer cells58. This suggests that the
changes in methylation marks at CpG island shores are
differentially regulated in HiPSCs and cancer cells58.
Another rather obvious tumorigenic risk in HiPSCs
is the use of integrating vectors for their derivation.
Although it has been demonstrated that the viral integra-
tion is not directly linked to the reprogramming proc-
ess59, the genomic alterations created, together with the
risk of reactivation of the introduced transgenes dur-
ing the propagation of the undifferentiated cells or on
At a glance
NATURE REVIEWS | CANCER
VOLUME 11 | APRIL 2011 | 269
© 2011 Macmillan Publishers Limited. All rights reserved
Chromosomal aberrations in HESCs
Chromosomal aberrations in HiPSCs
of somatic cells
their differentiation, pose a considerable threat that pre-
vents any clinical use of such genetically altered cells.
Therefore, a considerable amount of effort has recently
been put into producing virus-free and integration-free
mouse and human iPSCs using non-integrating adeno-
viruses60, expression plasmids61,62, episomal vectors63,
piggyBac transposition64, Cre-recombinase excisable
viruses65, direct delivery of reprogramming proteins66,67
and synthetic modified mRNAs68. This rapid improve-
ment in the technologies for the derivation of HiPSCs
is expected to reduce their tumorigenicity and improve
their safety, but a completely safe, simple and efficient
reprogramming method has yet to be developed.
Aneuploidy in HESCs and HiPSCs
The viral integration and the reactivation of the repro-
gramming factors are not the only possible sources for
genomic alterations in HiPSCs. Unlike HESCs, HiPSCs
are derived from mature somatic cells that have under-
gone multiple cell divisions and have lived long enough
to acquire genetic mutations. These mutations are
assumed to be random and rare, but if they confer pro-
liferative or anti-apoptotic advantages to the cells that
carry them, they might be selected for during the repro-
gramming process, which is an inefficient process that
results in HiPSCs with an abnormal genetic composition.
In addition, the reprogramming process is likely to
engage stress response pathways in the cells, and this
might encourage the accumulation of genetic changes in
the reprogrammed cells, regardless of the reprogramming
method used. Although most of these genetic changes
are expected to be selected against, and would therefore
disappear on the first few cell divisions in culture, some
might be advantageous for the cells and would there-
fore prevail. A recent study has provided evidence for
such chromosomal aberrations, resulting either from the
somatic cells of origin or from reprogramming stress,
leading to HiPSCs with aberrant karyotypes at early
passages69 (FIG. 1).
Much like HESCs, HiPSCs can be cultured in vitro for
long periods, and are therefore susceptible to acquiring
chromosomal aberrations, in what has become known as
culture adaptation (FIG. 1). Chromosomal aberrations in
HESCs are not acquired randomly during culture adap-
tation, as some are much more frequent than others. The
most common aberrations detected are gains of chromo-
somes 12 (especially 12p)6,70, 17 (in particular 17q)6,70,71,
20 (most often 20q11.21)19,71–73 and X6,74. Interestingly,
duplications of chromosomes 12, 17 and X are hallmarks
of germ cell tumours and embryonal carcinoma cells6,20,
and the 20q11.21 region is also amplified in various can-
cers75. Culture-adapted HESCs and HiPSCs often exhibit
increased proliferation capacity and decreased growth
factor dependence, and they often form more aggressive
malignant tumours when injected into mice17–20. Genes
that reside in the recurrent aberrant regions have been
suggested to contribute to the selective advantage that
allows the aberrant cells to survive and take over the cul-
ture6,75,76, but until recently no evidence had been provided
for the role of any specific genes in this process.
HiPSCs were assumed to be similar to HESCs in
regard to culture adaptation, but this issue had not
been comprehensively tested until recently. HiPSCs are
always karyotyped after their initial derivation, as part
of the required characterization for new pluripotent cell
lines, but their genomic stability after prolonged culture
had not been systematically examined. One attempt to
detect subkaryotypic alterations that were associated
with culture adaptation of three HiPSC lines resulted
in the suggestion that the genome of reprogrammed
cells is both normal and highly stable even after at least
However, we have recently conducted the first large-
scale analysis of aneuploidy in HiPSCs, inferring the
chromosomal structure of dozens of HiPSC lines from
their respective gene expression profiles69. Our study
demonstrated the duplication of chromosome 12 to be
the most common aberration in HiPSCs, and showed
that this trisomy rapidly takes over the population. It
also revealed that trisomy 12 is accompanied by overex-
pression of the important pluripotency factors NANOG
and growth/differentiation factor 3 (GDF3), providing
the first direct evidence for the involvement of specific
genes in the process of culture adaptation69. It remains
to be explored whether trisomy 12 also increases the
tumorigenicity of HiPSCs in vivo (this question is still
not resolved in HESCs77,78), and whether NANOG and
GDF3 have any role in this.
The genomic stability of HESCs is affected by
in vitro environmental conditions, such as feeder cells,
the make up of the culture medium, and the techniques
Figure 1 | Source of chromosomal aberrations in pluripotent stem cells. Human
embryonic stem cells (HESCs) and human induced pluripotent stem cells (HiPSCs) are
subjected to different selection pressures and thus might exhibit different types of
chromosomal aberrations at different rates. a | HESCs are derived from the inner cell
mass of blastocysts and mostly acquire chromosomal aberrations on prolonged growth
in culture. b | HiPSCs are derived from somatic cells that have undergone more cell
divisions and are thus more likely to be genetically abnormal. Some of these aberrations
might be selected for during the reprogramming process. The reprogramming process
itself is stressful for the cells and might introduce novel chromosomal aberrations, as well
as select for existing ones. Finally, HiPSCs also acquire chromosomal aberrations during
their prolonged passaging, and the techniques used for their culturing often differ from
those used for the growth of HESCs in vitro.
270 | APRIL 2011 | VOLUME 11
© 2011 Macmillan Publishers Limited. All rights reserved
used for cell passaging, freezing and thawing (reviewed
in REF. 75). Given the fact that HiPSCs are often more dif-
ficult to culture than HESCs, many laboratories imple-
ment slightly to moderately different techniques for the
culture of HESCs and HiPSCs. This could influence the
genomic instability of these cells and the selection pres-
sures they are subjected to, further contributing to the
potential difference in their tumorigenicity.
It is important to bear in mind that the genetic
changes that occur during the derivation of HiPSC lines
or during their adaptation in culture are not necessarily
limited to noticeable aneuploidies. Gains or losses of
small chromosomal regions that cannot be detected
using cytogenetic analyses, as well as point mutations
that cannot be detected by more sensitive DNA-based
methods (such as comparative genome hybridization
and copy number variation arrays), probably also occur.
Indeed, a recent study demonstrated a high frequency
of subchromosomal gains and deletions in HESCs and
HiPSCs, and showed that deleted regions often contain
tumour-suppressor genes, and that regions of chromo-
somal gain contain oncogenes79. Therefore, these minor
aberrations could have major functional implications for
the tumorigenicity of the HiPSCs. For example, a point
mutation that increases the proliferation of the cells, that
confers them with resistance to apoptosis or that hin-
ders their differentiation capacity, might result in more
aggressive teratomas. Indeed, the levels of expression
of tumour-suppressor genes have been found to affect
teratoma growth80. As advanced technologies, such as
whole-genome sequencing, become more accessible, we
predict that many such aberrations in HiPSCs will be
revealed. The causes for genetic instability in HiPSCs,
and potential ways to minimize it, have recently been
Epigenetic factors and HiPSC tumorigenicity
Differences in tumorigenicity between HESCs and
HiPSCs do not necessarily derive solely from genetic
differences, but might also be the consequence of epi-
genetic differences. A debate has recently taken place
regarding the claims that HiPSCs and HESCs system-
atically differ in their global gene expression patterns
and that extended culture brings them transcriptionally
closer26,82–84. Such epigenetic differences may be relevant
to the evaluation of the tumorigenicity of HiPSCs. First,
these differentially expressed genes might include onco-
genes and tumour suppressors that directly affect the
tumorigenicity of the cells. Second, if HiPSCs become
transcriptionally closer to HESCs during extended cul-
ture this would create an interesting dilemma: on the
one hand, clinical use of HiPSCs would require them to
be as transcriptionally similar to HESCs as possible; on
the other hand, extended culture often leads to chromo-
somal aberrations (as described above), and thus might
increase the tumorigenicity of the cells.
Consistent with the idea of differences in gene
expression between HESCs and HiPSCs, a recent study
compared the miRNA expression of two HiPSCs with
the miRNA expression of four HESC lines, and identi-
fied ten cancer-related miRNAs that are overexpressed
(> tenfold difference) in HiPSCs85. Another study has
recently reported cancer-related epigenetic abnormali-
ties, such as alterations in cancer-specific gene promoter
DNA methylation, that arise early during reprogram-
ming and that persist in HiPSCs86. Thus, accumulating
data suggest that the reprogramming process is often
accompanied not only by genetic abnormalities, but also
by epigenetic alterations, which are expected to increase
the tumorigenicity of HiPSCs.
In line with the claim that early passage HiPSCs dis-
play unique transcriptional programmes that are attenu-
ated on continuous passaging, recent reports have shown
that iPSCs retain epigenetic memory of their cells of
origin, both in mouse28,29 and in human27,87,88. This epi-
genetic memory manifests as differential gene expres-
sion and altered differentiation capacity, but it would be
interesting to further explore whether it also affects the
tumorigenicity of the cells, owing to one of the following
possibilities: memory of cellular epigenetic transforma-
tions (that is, a direct effect on tumorigenicity) and/or
memory of epigenetic characteristics that make some
types of cells more vulnerable to transformation (that
is, an indirect effect on tumorigenicity). The memory
of cellular epigenetic transformations has already been
demonstrated in a study that carried out a large-scale
methylation profile of somatic donor cells and identified
aberrant methylation at hundreds of sites, only some of
which were reversed following reprogramming89. Such
epigenetic memory might result in tumorigenic dif-
ferences between HiPSC lines from different somatic
sources. Interestingly, in mice the epigenetic memory
is gradually lost during extended culture of iPSCs or
during further cycles of reprogramming 28,29, and if this
finding also applies to HiPSCs, it might contribute to
tumorigenic differences between early and late-passage
cells of the same HiPSC line.
Another important source for potential epigenetic
differences between HESCs and HiPSCs is the status of
genomic imprinting. Aberrant imprinting is associated
with neoplasia and is evident in some types of human
cancer (reviewed in REFS 90,91). Aberrant silencing or
activation of imprinted genes during the reprogram-
ming process might have implications for both their
differentiation capacity and their tumorigenicity. Recent
studies found variability in the expression of imprinted
genes among different lines of both mouse and human
iPSCs30,92. Although HESCs exhibit fairly stable genomic
imprinting at early passages93–95, some HiPSCs show
aberrant expression of imprinted genes30. Furthermore,
the expression state of a single imprinted gene cluster
was recently shown to affect the developmental potential
of mouse iPSCs and to distinguish them from MESCs92.
Thus, it is reasonable to assume that aberrant expression
of imprinted genes would also affect the tumorigenicity of
some HiPSC lines.
Aberrant imprinting might also occur during the
growth of pluripotent stem cells in culture, and its
acquisition might therefore be defined as an epigenetic
cellular adaptation. It was found that in vitro culture
over extended periods affects the integrity of imprinted
gene expression in HESCs93,96, and it was suggested that
NATURE REVIEWS | CANCER
VOLUME 11 | APRIL 2011 | 271
© 2011 Macmillan Publishers Limited. All rights reserved
culturing practices result in changes of DNA methy-
lation at differentially methylated regions (DMRs)96.
Imprinting aberrations are also likely to occur as a result
of similar culture techniques used during the prolonged
culture of HiPSCs. As the imprinting in HiPSCs is less
stable initially, one might anticipate that these aber-
rations would be even more substantial in these cells.
However, the nature and severity of acquired imprinting
aberrations in HiPSCs remain to be studied.
Cell cycle regulation in HESCs and HiPSCs
The remarkable self-renewal capabilities of human
pluripotent stem cells require a unique cell cycle regu-
lation, which allows them to divide independently of
extrinsic mitogenic signals. HESCs and HiPSCs are
characterized by abbreviated gap phases, and a high
proportion of cells in the S phase and the M phase
of the cell cycle97. The role of cell cycle regulation in
pluripotency and reprogramming has been reviewed
The structure and regulation of the cell cycle influ-
ence the tumorigenicity of HESCs and HiPSCs, as self-
renewal is a determinant for their tumorigenic potential.
Although many of the genes that regulate pluripotency
and the cell cycle are involved in maintaining the self-
renewal abilities of these cells26,99, MYC activity has been
specifically shown to account for many of their cell cycle
properties98. Other important cell cycle regulators, such
as p53 and cyclin D1, were also found to be intrinsically
involved in the reprogramming process100.
Cell cycle regulation probably has an important
role in the propensity of HESCs and HiPSCs to acquire
chromosomal aberrations. A recent report linked the
unique cell cycle of HESCs to numerical centrosomal
abnormalities during mitosis101, which might account, at
least partly, for their enhanced chromosomal instability,
and thus increase their tumorigenicity. Supporting this
idea, another recent report found that frequently aber-
rant chromosomal regions in HiPSCs are enriched for
cell cycle-related genes69.
To date, no report has described the differences in
cell cycle regulation between HESCs and HiPSCs, and
HiPSCs have been reported to acquire the distinct cell
cycle properties of HESCs during reprogramming97.
However, the risk for reactivation of the reprogramming
factors (especially MYC), as well as the risk for acquir-
ing genetic abnormalities early on during propagation
in culture, might make HiPSCs more vulnerable for cell
cycle aberrations. Further studies are needed to deter-
mine whether such differences in cell cycle regulation
sometimes exist, and whether these potential differences
affect the tumorigenicity of the cells.
HiPSC tumorigenicity: are the concerns relevant?
The use of potent oncogenes for the process of repro-
gramming, the involvement of the p53 pathway as a
barrier against this process and the epigenetic memory
of HiPSCs, all raise a possible risk that probably does
not exist in HESCs: namely, the risk for the develop-
ment of somatic tumours (rather than the development
of teratomas or teratocarcinomas). Indeed, the original
generation of germline-competent mouse iPSCs resulted
in various types of tumours owing to the reactivation of
Myc102, and many of the four-gene iPSC mice chimaeras
died from cancer within the first few months of their
lives103. Therefore, the risk for the formation of somatic
tumours is not merely hypothetical.
As mentioned above, although HESCs are subjected
to selection in culture, which often results in genomic
instability, HiPSCs are not only subjected to the same
selection pressure but also to additional selection pres-
sures during the reprogramming process. These stronger
selection pressures might increase chromosomal insta-
bility (as is evident from the appearance of chromosomal
aberrations early on during their growth in culture69)
and result in the formation of more aggressive terato-
mas, or even teratocarcinomas. In agreement, some
reprogrammed lines fail to differentiate and form undif-
ferentiated tumours on injection into immunodeficient
mice104. The potential differences in the tumorigenicity
of HESCs and HiPSCs, and their possible consequences
for tumour formation, are summarized in TABLE 1 and
shown in FIG. 2.
A study of the tumorigenicity of mouse iPSCs has
recently been published105. The authors compared
iPSCs from different origins (mouse embryonic
fibroblasts (MEFs) and several adult tissues), with or
without Myc retroviral transduction, and with or with-
out drug selection for the expression of pluri potency
genes (Nanog and Fbxo15). These various iPSCs were
also differentiated into secondary neurospheres (SNS)
and were examined for their teratoma-forming pro-
pensity after injection into the brains of non-obese
diabetic-SCID mice. Although the teratoma-formation
capability of SNS from MEF-derived iPSCs was similar
to that of SNS from embryonic stem cells, SNS from
iPSCs that were derived from different adult tissues
varied substantially in their teratoma-forming pro-
pensity, and some of them formed more aggressive,
undifferentiated teratocarcinomas. The aggressive-
ness of the tumours was found to correlate with the
number of residual pluripotent stem cells in the SNS.
Surprisingly, the use of the Myc retrovirus and the
presence of drug selection were not found to affect
the teratoma-formation capability of the iPSC-derived
neurospheres105. Although these results highlight the
potential differences in the tumorigenicity of iPSCs
from different somatic sources, consistent with the epi-
genetic memory studies described above, the genetic
composition of the iPSCs that were used in this study
has not been reported. Therefore, the difference in
tumorigenicity between the different lines might be a
consequence of chromosomal aberrations, rather than
of the tissue of origin.
A study80 that compared the tumorigenicity of retro-
virally derived and transgene-free HiPSCs observed no
substantial tumorigenic differences between the two
groups, as judged by blood microvessel density (MVD)
within the teratomas that formed80,106. Interestingly,
a correlation was evident between the MVD and the
level of expression of p21 and p53 in the HiPSC lines.
These results again suggest that the p53 pathway and the
272 | APRIL 2011 | VOLUME 11
© 2011 Macmillan Publishers Limited. All rights reserved
reprogramming process are linked, and that this path-
way might affect the aggressiveness of HiPSC-derived
More recently, direct comparisons of teratoma for-
mation by HESCs and HiPSCs have been reported86,107.
One study found that HiPSCs develop teratomas more
efficiently and faster than HESCs, regardless of the site of
injection. No differences in the composition of the terato-
mas were observed107. Although a karyotype analysis was
conducted in this study, ruling out major chromosomal
aberration as an explanation of the observed differences,
smaller genomic aberrations that can be detected only by
using higher-resolution methods, could still account for
these results. Furthermore, the comparison of teratoma
composition was carried out by staining for the three
germ layers only, thus not precluding the possibility that
HiPSC-derived teratomas contain undifferentiated foci
and are more aggressive than HESC-derived teratomas.
Indeed, a second study that compared teratomas that
were formed by HiPSC lines with a normal karyotype
with teratomas that were formed by HESC lines reported
that all of the HiPSC-derived teratomas examined,
although none of the HESC-derived teratomas, con-
tained malignant characteristics (such as, focal necrosis,
nuclear polymorphism, high mitotic rates and infiltration
into the mouse musculature)86. A more comprehensive,
higher-resolution comparison of teratoma formation by
HESCs and HiPSCs is therefore warranted.
Coping with the tumorigenicity risks
In order to develop safe HESC- and HiPSC-derived
treatments, the tumorigenicity hurdle must be over-
come, and much research has been dedicated in recent
years to this purpose. The various strategies that might
cope with the tumorigenicity risk have been reviewed
elsewhere2,3,108, and can generally be divided into three
categories3: terminal differentiation or complete elimi-
nation of residual pluripotent stem cells from culture;
interfering with tumour-progression genes to prevent
tumour formation from the residual pluripotent cells;
and tumour detection and elimination on formation in
the patient’s body.
In view of reports suggesting that as few as several
hundred pluripotent cells are sufficient to generate
tumours109,110, there is no doubt that the safest HESC- and
HiPSC-derived treatments would require a 100% pure
population of differentiated cells, and this can be achieved
either by complete differentiation or by complete ablation
of undifferentiated cells from mixed populations. Various
attempts to achieve this goal have been reported108,111, with
the most recent advances being the targeted elimination of
pluripotent cells by cytotoxic antibodies112,113, and the sep-
aration of undifferentiated HESCs from a heterogeneous
cell population on the basis of pluripotent-specific cell
surface molecules, using magnetic-activated cell sort-
ing (MACS) and fluorescence-activated cell sorting
(FACS)114. However, none of the current methods results
Table 1 | Comparison of the tumorigenicity between HESCs and HiPSCs
Cell of origin
ICM cells that have undergone
very few divisions*
•?Mature somatic cells that have undergone many cell
divisions and have been more exposed to genetic and
•?Might result in mutations and/or aberrations of somatic
Derivation process A relatively minor selection
•?A major selection pressure owing to forced drastic
change of epigenetic landscape‡
•?Might result in mutations and/or aberrations owing to
Viral integration Not applicableMost of the current methods still use viral vectors for
Not applicableCurrent methods upregulate oncogenes in the
Cellular adaptation to
Prolonged growth in culture
often results in gains of
chromosomes 12, 17, 20 and X‡
Prolonged growth in culture often results in gains of
Cell of originSimilarity of global gene
expression with some cancers
(onco-fetal genes are highly
•?Similarity of global gene expression with some cancers
(onco-fetal genes are highly expressed) ‡
•?Epigenetic memory of somatic transformations and/or of
susceptible traits of the somatic tissue‡
Derivation process No substantial epigenetic
aberrations are known to occur
in the process*
•?Cancer-related epigenetic abnormalities arise during
•?Relaxation of imprinting might also occur in the process‡
Cellular adaptation to
Relaxation of imprinting might
occur in culture*
Relaxation of imprinting might occur in culture*
HESCs, human embryonic stem cells; HiPSCs, human induced pluripotent stem cells; ICM, inner cell mass. *Medium risk of tumour
generation. ‡High risk of tumour generation.
NATURE REVIEWS | CANCER
VOLUME 11 | APRIL 2011 | 273
© 2011 Macmillan Publishers Limited. All rights reserved
Nature Reviews | Cancer
Genetic aberrations due
to culture adaptation
Genetic and epigenetic
aberrations due to
and culture adaptation
Genetic and epigenetic
in a 100% pure differentiated culture, so that complete
elimination of all differentiated cells from a mixed culture
might require more robust measures.
Given the current limited ability to produce a pure
population of differentiated HESC or HiPSC progeny, an
alternative method could be to interfere with genes that are
important for teratoma formation, but are dispensable for
mature tissues20. It was recently found that the expression
of survivin (encoded by BIRC5), a classical onco-fetal gene
with anti-apoptotic activity, is enriched in undifferentiated
HESCs and their derived teratomas115. The ablation of sur-
vivin expression, both genetically and pharmacologically,
stimulated apoptosis in cultured HESCs and in their ter-
atomas115. Thus, the discovery of more teratoma-associated
genes and ways to molecularly target them is a promising
strategy to cope with the tumorigenicity of these cells.
A similar approach would be to interfere with genes
that are essential for self-renewal in human pluripotent
stem cells, thus forcing them to differentiate and decreas-
ing their tumorigenic potential. The most obvious candi-
date for such an approach would be endo genous MYC, as
several studies have recently shown that MYC represses
differentiation and maintains the self-renewal of mouse
and human pluripotent stem cells116,117 (reviewed in
REF. 118). The potential of MYC ablation for teratoma
prevention seems even more promising in view of a
recent report that MYC regulatory networks account for
most of the transcriptional similarity between embry-
onic stem cells and cancer cells119. However, interference
with MYC or with other pluripotency genes will inevi-
tably compromise the pluripotency and the self-renewal
of the cells, and will thus need to be restricted to the
end point-differentiated cultures, in which the residual
pluripotent stem cells are undesired.
The third, albeit problematic, strategy to cope with
the tumorigenicity of pluripotent cells is to detect and
attack the tumours once their formation has already
been initiated. The genetic introduction of ‘suicide’ genes
into HESCs is one potential way to eliminate teratomas
after their formation, using specific drugs120. Another
previously suggested idea, limited to specific diseases
such as diabetes, is to make these tumours irrelevant by
encapsulating the HESC-derived grafts, thus enabling
their safe removal and preventing them from spreading
through the patient’s body121–123. These techniques pro-
vide another layer of defence against potential tumours,
but as they do not prevent the tumour formation itself,
they are doomed to remain complementary, rather than
primary, coping strategies.
In principle, the same coping strategies that apply for
HESCs should also apply for HiPSCs. However, the dif-
ferences that might exist between HESCs and HiPSCs in
their global gene expression pattern and epigenetic land-
scape require that every specific separation, purification
or interference method described above needs to be
tested on HiPSCs in order to confirm their relevance for
HiPSC-based therapies. This validation is of the utmost
importance given the risk that HiPSCs might generate
somatic tumours. Intervening with teratoma-specific
genes, for example, will not necessarily prevent the
formation of somatic tumours by residual HiPSCs, as dif-
ferent molecular mechanisms underlie the development
of different kinds of tumours.
The potential increased aggressiveness of HiPSCs,
which might result in the formation of somatic tumours,
aggressive teratomas or teratocarcinomas, requires the
implementation of strict safety measures. Constant
genetic and epigenetic safety estimations need to be rou-
tinely carried out if these cells are to be used in the clinic.
In order to monitor for genetic transformations that
might be associated with the reprogramming process or
with the culture of HiPSCs, more sensitive measures will
be needed. Although integration-, viral- and oncogene-
free reprogramming is expected to decrease genetic
abnormalities, a routine analysis of the DNA content of
HiPSCs will nonetheless be required. High-resolution
DNA analysis of regions that are prone to aberrations
could complement the standard karyotype analysis;
whole-genome sequencing of the cell lines might also be
required in some cases. To monitor for epigenetic trans-
formations that might arise in HiPSCs, it will be impor-
tant to verify the loss of undesired epigenetic memory,
as well as the normal expression levels of the imprinted
genes and of the known oncogenes. This can be achieved
by global and specific gene expression analyses, com-
plemented by an analysis of epigenetic markers, such as
DNA methylation, on susceptible genes.
Figure 2 | Tumour formation in pluripotent stem cells. Residual diploid human
embryonic stem cells (HESCs) that have not undergone differentiation are expected to
form benign teratomas when transplanted into human patients. Although some
aneuploid cells are also expected to form benign teratomas, other aneuploid cells
might undergo transformation and form malignant teratocarcinomas. Residual diploid
human induced pluripotent stem cells (HiPSCs), which are similar HESCs, are also
expected to result in benign teratomas. However, some diploid HiPSCs that appear to
be genetically normal might form teratocarcinomas owing to epigenetic
transformation. This epigenetic transformation might result from epigenetic memory
(tissue-of-origin traits and epigenetic alterations of somatic origin) and/or from the
reprogramming process itself (such as the reactivation of oncogenes involved in
reprogramming and epigenetic alterations introduced during reprogramming).
Aneuploid HiPSCs might also form teratocarcinomas and genetic aberrations might
occur earlier than in HESCs owing to somatic mutations, reprogramming stress, viral
integration and/or culture adaptation. Unlike HESCs, diploid and aneuploid HiPSCs
also harbour a risk for the formation of somatic tumours, owing to genetic and
epigenetic transformation that is acquired in the somatic tissue of origin or during
reprogramming. Dashed arrows indicate possible outcomes; solid arrows indicate
definite outcomes. The image of HESCs is adapted from REF. 126.
274 | APRIL 2011 | VOLUME 11
© 2011 Macmillan Publishers Limited. All rights reserved
The concerns and possibilities raised by the genera-
tion and use of HiPSCs will eventually require further
clarification if optimal HiPSCs are to be defined. For
example, if prolonged growth in culture results in HiPSCs
acquiring chromosomal aberrations69, but also results in
an increasing similarity to their HESC counterparts26,82
and the loss of the markers of their source of origin28,29,
a need to determine the desirable ‘duration-in-culture’ of
HiPSCs arises. In order to do this, there is a need for sys-
tematic research that would define the optimal time-point
(if such a point even exists) when the similarity to HESCs
is already maximal but the epigenetic memory and the
tumorigenicity are minimal. Alternatively, improving the
methods of HiPSC derivation in ways that more firmly
erase the epigenetic memory of the cells would help
reconcile these conflicting tumorigenic forces.
Moreover, it might be that the most potent HiPSCs
— the ones that appear in culture first, grow the fastest
and/or resemble HESCs the most — are not the most
suitable candidates for therapeutic purposes. Partially
reprogrammed cells, or fully reprogrammed cells that
grow more slowly in culture, might be better for clinical
application (as long as they can be differentiated into the
required cell type). In addition, HiPSCs produced from
embryonic tissues, such as cord blood, may minimize
the acquisition of genetic and epigenetic mutations, and
thus may be safer than HiPSCs that are derived from
adult somatic cells124. This might also be true for HiPSCs
that are derived from stem cells. These cells have been
reported to require the transduction of only one (for
neural stem cells125) or two (for cord blood stem cells124)
reprogramming factors, and consequently decrease the
risk for the reactivation of these oncogenes. Last, dif-
ferences in the genetic backgrounds and epigenetic
signatures between individual HiPSC lines might make
specific lines more tumorigenic than others, regardless
of their cellular origin, derivation process and cell cul-
ture conditions. The tumorigenicity of the cells is there-
fore expected to have a more central role in defining the
optimal HiPSC lines for regenerative medicine.
Although much effort has been put into trying to generate
integration- and virus-free HiPSCs, and into devel-
oping new and improved coping strategies with the
tumorigenicity of pluripotent stem cells, most of the
mechanistic biological research in the field currently
focuses on the generation of HiPSCs that are as similar
as possible to HESCs, and on studying the emerging dif-
ferences between these two cell types. As the reprogram-
ming field matures and moves towards the clinic, more
biological research on the tumorigenicity of HiPSCs, its
sources and its consequences, should be forthcoming.
Once we better understand the biolgy of cellular repro-
gramming and the biological differences between HESCs
and HiPSCs, we should be able to generate HiPSCs that
are only functionally similar to HESCs but are less
tumorigenic and therefore safer for clinical application.
Although the data reviewed here might suggest that
HiPSCs are more tumorigenic than HESCs, their somatic
origin might in fact turn out to be advantageous in this
regard. If we succeed in reprogramming cells so that
they acquire only desirable aspects of cellular potency
and can differentiate in vitro into functional cells of the
type required, but at the same time lack the ability to
form tumours in vivo, this might enable us to disentangle
the Gordian knot that links pluripotency and tumori-
genicity. As our control over the epigenetic landscape of
the cell increases, this dream might not be as far from
realization as it was just a few years ago.
Note added in proof
While this article was in the press, three important arti-
cles that further stress the genetic and epigenetic differ-
ences between human embryonic stem cells (HESCs) and
human induced pluripotent stem cells (HiPSCs) were
published. Using high-resolution genetic approaches,
two of these papers reported copy number variations
(CNVs)127 and protein-coding point mutations128 that
arise during the reprogramming process. The point
mutations were found to be enriched in cancer-related
genes128. In the third paper, a whole-genome profiling of
DNA methylation was carried out at single-base resolu-
tion, revealing somatic epigenetic memory and aber-
rant reprogramming of DNA methylation in HiPSCs129.
Together, these papers suggest genetic and epigenetic
abnormalities that distinguish HiPSCs from HESCs,
and strongly support the idea, which was raised in this
Review, that HiPSCs are likely to be more tumorigenic
Dreesen, O. & Brivanlou, A. H. Signaling pathways in
cancer and embryonic stem cells. Stem Cell Rev. 3,
Knoepfler, P. S. Deconstructing stem cell
tumorigenicity: a roadmap to safe regenerative
medicine. Stem Cells 27, 1050–1056 (2009).
Blum, B. & Benvenisty, N. The tumorigenicity of
human embryonic stem cells. Adv. Cancer Res. 100,
References 2 and 3 are recent reviews on the
tumorigenicity of HESCs and possible coping
Amit, M. et al. Clonally derived human embryonic
stem cell lines maintain pluripotency and proliferative
potential for prolonged periods of culture. Dev. Biol.
227, 271–278 (2000).
Thomson, J. A. et al. Embryonic stem cell lines derived
from human blastocysts. Science 282, 1145–1147
Baker, D. E. et al. Adaptation to culture of human
embryonic stem cells and oncogenesis in vivo. Nature
Biotech. 25, 207–215 (2007).
Thorough overview of the chromosomal aberrations
observed in HESCs in culture.
Harrison, N. J., Baker, D. & Andrews, P. W. Culture
adaptation of embryonic stem cells echoes germ cell
malignancy. Int. J. Androl. 30, 275–281 (2007).
Hiyama, E. & Hiyama, K. Telomere and telomerase
in stem cells. Br. J. Cancer 96, 1020–1024 (2007).
Eilers, M. & Eisenman, R. N. Myc’s broad reach. Genes
Dev. 22, 2755–2766 (2008).
10. Evans, P. M. & Liu, C. Roles of Krupel-like factor 4 in
normal homeostasis, cancer and stem cells. Acta
Biochim. Biophys. Sin. 40, 554–564 (2008).
11. Sperger, J. M. et al. Gene expression patterns in
human embryonic stem cells and human pluripotent
germ cell tumors. Proc. Natl Acad. Sci. USA 100,
12. Ben-Porath, I. et al. An embryonic stem cell-like gene
expression signature in poorly differentiated aggressive
human tumors. Nature Genet. 40, 499–507 (2008).
13. Wong, D. J. et al. Module map of stem cell genes
guides creation of epithelial cancer stem cells. Cell
Stem Cell 2, 333–344 (2008).
14. Neveu, P. et al. MicroRNA profiling reveals two distinct
p53-related human pluripotent stem cell states. Cell
Stem Cell 7, 671–681 (2010).
15. Calvanese, V. et al. Cancer genes hypermethylated in
human embryonic stem cells. PLoS ONE 3, e3294
16. Bjorklund, L. M. et al. Embryonic stem cells develop
into functional dopaminergic neurons after
transplantation in a Parkinson rat model. Proc. Natl
Acad. Sci. USA 99, 2344–2349 (2002).
17. Herszfeld, D. et al. CD30 is a survival factor
and a biomarker for transformed human pluripotent
stem cells. Nature Biotech. 24, 351–357 (2006).
NATURE REVIEWS | CANCER
VOLUME 11 | APRIL 2011 | 275
© 2011 Macmillan Publishers Limited. All rights reserved
18. Yang, S. et al. Tumor progression of culture-adapted
human embryonic stem cells during long-term
culture. Genes Chromosomes Cancer 47, 665–679
19. Werbowetski-Ogilvie, T. E. et al. Characterization of
human embryonic stem cells with features of
neoplastic progression. Nature Biotech. 27, 91–97
First demonstration of altered proliferation and
differentiation capacities in adapted HESC lines
with subkaryotypic genetic abnormalities.
20. Blum, B. & Benvenisty, N. The tumorigenicity of
diploid and aneuploid human pluripotent stem cells.
Cell Cycle 8, 3822–3830 (2009).
Recent perspective of the tumorigenicity of HESCs
and its relationship with genomic instability of
HESCs in culture.
21. Martin, G. R. Isolation of a pluripotent cell line from
early mouse embryos cultured in medium conditioned
by teratocarcinoma stem cells. Proc. Natl Acad. Sci.
USA 78, 7634–7638 (1981).
22. Shih, C. C., Forman, S. J., Chu, P. & Slovak, M. Human
embryonic stem cells are prone to generate primitive,
undifferentiated tumors in engrafted human fetal
tissues in severe combined immunodeficient mice.
Stem Cells Dev. 16, 893–902 (2007).
23. Takahashi, K. & Yamanaka, S. Induction of pluripotent
stem cells from mouse embryonic and adult fibroblast
cultures by defined factors. Cell 126, 663–676
24. Takahashi, K. et al. Induction of pluripotent stem cells
from adult human fibroblasts by defined factors. Cell
131, 861–872 (2007).
25. Yu, J. et al. Induced pluripotent stem cell lines derived
from human somatic cells. Science 318, 1917–1920
26. Chin, M. H. et al. Induced pluripotent stem cells and
embryonic stem cells are distinguished by gene
expression signatures. Cell Stem Cell 5, 111–123
First study describing global gene expression
differences between HESCs and HiPSCs.
27. Urbach, A., Bar-Nur, O., Daley, G. Q. & Benvenisty, N.
Differential modeling of fragile X syndrome by human
embryonic stem cells and induced pluripotent stem
cells. Cell Stem Cell 6, 407–411 (2010).
28. Polo, J. M. et al. Cell type of origin influences the
molecular and functional properties of mouse induced
pluripotent stem cells. Nature Biotech. 28, 848–855
29. Kim, K. et al. Epigenetic memory in induced
pluripotent stem cells. Nature 467, 285–290 (2010).
References 28 and 29 are comprehensive studies
describing epigenetic memory in mouse iPSCs.
30. Pick, M. et al. Clone- and gene-specific aberrations of
parental imprinting in human induced pluripotent
stem cells. Stem Cells 27, 2686–2690 (2009).
31. Albihn, A., Johnsen, J. I. & Henriksson, M. A. MYC in
oncogenesis and as a target for cancer therapies. Adv.
Cancer Res. 107, 163–224 (2010).
32. Ruggero, D. The role of Myc-induced protein synthesis
in cancer. Cancer Res. 69, 8839–8843 (2009).
33. Tian, Y. et al. MicroRNA-10b promotes migration
and invasion through KLF4 in human esophageal
cancer cell lines. J. Biol. Chem. 285, 7986–7994
34. Lambertini, C., Pantano, S. & Dotto, G. P. Differential
control of Notch1 gene transcription by Klf4 and Sp3
transcription factors in normal versus cancer-derived
keratinocytes. PLoS ONE 5, e10369 (2010).
35. Rageul, J. et al. KLF4-dependent, PPARγ-induced
expression of GPA33 in colon cancer cell lines. Int.
J. Cancer 125, 2802–2809 (2009).
36. Asadi, M. H. et al. OCT4B1, a novel spliced variant of
OCT4, is highly expressed in gastric cancer and acts as
an anti-apoptotic factor. Int. J. Cancer 3 Nov 2010
37. Wang, Y. et al. Oct-4B isoform is differentially
expressed in breast cancer cells: hypermethylation of
regulatory elements of Oct-4A suggests an alternative
promoter and transcriptional start site for Oct-4B
transcription. Biosci. Rep. 31, 109–115 (2010).
38. Peng, S., Maihle, N. J. & Huang, Y. Pluripotency
factors Lin28 and Oct4 identify a sub-population of
stem cell-like cells in ovarian cancer. Oncogene 29,
39. Sholl, L. M., Barletta, J. A., Yeap, B. Y., Chirieac, L. R.
& Hornick, J. L. Sox2 protein expression is an
independent poor prognostic indicator in stage I lung
adenocarcinoma. Am. J. Surg. Pathol. 34, 1193–1198
40. Ji, J. & Zheng, P. S. Expression of Sox2 in human
cervical carcinogenesis. Hum. Pathol. 41, 1438–1447
41. Schoenhals, M. et al. Embryonic stem cell markers
expression in cancers. Biochem. Biophys. Res.
Commun. 383, 157–162 (2009).
42. Ramos-Mejia, V., Munoz-Lopez, M., Garcia-Perez, J. L.
& Menendez, P. iPSC lines that do not silence the
expression of the ectopic reprogramming factors may
display enhanced propensity to genomic instability.
Cell Res. 20, 1092–1095 (2010).
43. Nakagawa, M. et al. Generation of induced pluripotent
stem cells without Myc from mouse and human
fibroblasts. Nature Biotech. 26, 101–106 (2008).
44. Nakagawa, M., Takizawa, N., Narita, M., Ichisaka, T. &
Yamanaka, S. Promotion of direct reprogramming by
transformation-deficient Myc. Proc. Natl Acad. Sci.
USA 107, 14152–14157 (2010).
45. Li, W. et al. Generation of human-induced pluripotent
stem cells in the absence of exogenous Sox2. Stem
Cells 27, 2992–3000 (2009).
46. Li, W. et al. Generation of rat and human induced
pluripotent stem cells by combining genetic
reprogramming and chemical inhibitors. Cell Stem Cell
4, 16–19 (2009).
47. Huangfu, D. et al. Induction of pluripotent stem cells
from primary human fibroblasts with only Oct4 and
Sox2. Nature Biotech. 26, 1269–1275 (2008).
48. Zhu, S. et al. Reprogramming of human primary
somatic cells by OCT4 and chemical compounds. Cell
Stem Cell 7, 651–655 (2010).
49. Marion, R. M. et al. A p53-mediated DNA damage
response limits reprogramming to ensure iPS cell
genomic integrity. Nature 460, 1149–1153 (2009).
50. Utikal, J. et al. Immortalization eliminates a roadblock
during cellular reprogramming into iPS cells. Nature
460, 1145–1148 (2009).
51. Hong, H. et al. Suppression of induced pluripotent
stem cell generation by the p53-p21 pathway. Nature
460, 1132–1135 (2009).
52. Li, H. et al. The Ink4/Arf locus is a barrier for iPS cell
reprogramming. Nature 460, 1136–1139 (2009).
53. Kawamura, T. et al. Linking the p53 tumour
suppressor pathway to somatic cell reprogramming.
Nature 460, 1140–1144 (2009).
54. Menendez, S., Camus, S. & Belmonte, J. C. p53:
guardian of reprogramming. Cell Cycle 9, 3887–3891
55. Marion, R. M. et al. Telomeres acquire embryonic
stem cell characteristics in induced pluripotent stem
cells. Cell Stem Cell 4, 141–154 (2009).
56. Yehezkel, S. et al. Reprogramming of telomeric regions
during the generation of human induced pluripotent
stem cells and subsequent differentiation into fibroblast-
like derivatives. Epigenetics 6, 63–75 (2011).
57. Vera, E., Canela, A., Fraga, M. F., Esteller, M. &
Blasco, M. A. Epigenetic regulation of telomeres in
human cancer. Oncogene 27, 6817–6833 (2008).
58. Doi, A. et al. Differential methylation of tissue- and
cancer-specific CpG island shores distinguishes human
induced pluripotent stem cells, embryonic stem cells
and fibroblasts. Nature Genet. 41, 1350–1353 (2009).
59. Aoi, T. et al. Generation of pluripotent stem cells from
adult mouse liver and stomach cells. Science 321,
60. Stadtfeld, M., Nagaya, M., Utikal, J., Weir, G. &
Hochedlinger, K. Induced pluripotent stem cells
generated without viral integration. Science 322,
61. Okita, K., Nakagawa, M., Hyenjong, H., Ichisaka, T. &
Yamanaka, S. Generation of mouse induced
pluripotent stem cells without viral vectors. Science
322, 949–953 (2008).
62. Kaji, K. et al. Virus-free induction of pluripotency and
subsequent excision of reprogramming factors. Nature
458, 771–775 (2009).
63. Yu, J. et al. Human induced pluripotent stem cells free
of vector and transgene sequences. Science 324,
64. Woltjen, K. et al. piggyBac transposition reprograms
fibroblasts to induced pluripotent stem cells. Nature
458, 766–770 (2009).
65. Soldner, F. et al. Parkinson’s disease patient-derived
induced pluripotent stem cells free of viral
reprogramming factors. Cell 136, 964–977 (2009).
66. Zhou, H. et al. Generation of induced pluripotent stem
cells using recombinant proteins. Cell Stem Cell 4,
67. Kim, D. et al. Generation of human induced pluripotent
stem cells by direct delivery of reprogramming
proteins. Cell Stem Cell 4, 472–476 (2009).
68. Warren, L. et al. Highly efficient reprogramming to
pluripotency and directed differentiation of human
cells with synthetic modified mRNA. Cell Stem Cell 7,
69. Mayshar, Y. et al. Identification and classification of
chromosomal aberrations in human induced
pluripotent stem cells. Cell Stem Cell 7, 521–531
First comprehensive study of chromosomal
aberrations observed in HiPSCs in culture.
70. Draper, J. S., Moore, H. D., Ruban, L. N., Gokhale,
P. J. & Andrews, P. W. Culture and characterization of
human embryonic stem cells. Stem Cells Dev. 13,
71. Maitra, A. et al. Genomic alterations in cultured
human embryonic stem cells. Nature Genet. 37,
72. Lefort, N. et al. Human embryonic stem cells reveal
recurrent genomic instability at 20q11.21. Nature
Biotech. 26, 1364–1366 (2008).
73. Spits, C. et al. Recurrent chromosomal abnormalities
in human embryonic stem cells. Nature Biotech. 26,
74. Inzunza, J. et al. Comparative genomic hybridization
and karyotyping of human embryonic stem cells
reveals the occurrence of an isodicentric
X chromosome after long-term cultivation. Mol. Hum.
Reprod. 10, 461–466 (2004).
75. Lefort, N., Perrier, A. L., Laabi, Y., Varela, C. &
Peschanski, M. Human embryonic stem cells and
genomic instability. Regen. Med. 4, 899–909 (2009).
Recent review on the genomic instability of HESCs.
76. Narva, E. et al. High-resolution DNA analysis of
human embryonic stem cell lines reveals culture-
induced copy number changes and loss of
heterozygosity. Nature Biotech. 28, 371–377 (2010).
77. Gertow, K. et al. Trisomy 12 in HESC leads to no
selective in vivo growth advantage in teratomas, but
induces an increased abundance of renal development.
J. Cell. Biochem. 100, 1518–1525 (2007).
78. Moon, S. H. et al. Effect of chromosome instability on
the maintenance and differentiation of human
embryonic stem cells in vitro and in vivo. Stem Cell
Res. 6, 50–59 (2011).
79. Laurent, L. C. et al. Dynamic changes in the copy
number of pluripotency and cell proliferation genes in
human ESCs and iPSCs during reprogramming and
time in culture. Cell Stem Cell 8, 106–118 (2011).
80. Moriguchi, H., Chung, R. T. & Sato, C. Tumorigenicity
of human induced pluripotent stem cells depends on
the balance of gene expression between p21 and p53.
Hepatology 51, 1088–1089 (2010).
81. Ben-David, U., Benvenisty, N. & Mayshar, Y. Genetic
instability in human induced pluripotent stem cells:
classification of causes and possible safeguards. Cell
Cycle 9, 4603–4604 (2010).
82. Chin, M. H., Pellegrini, M., Plath, K. & Lowry, W. E.
Molecular analyses of human induced pluripotent
stem cells and embryonic stem cells. Cell Stem Cell 7,
83. Guenther, M. G. et al. Chromatin structure and gene
expression programs of human embryonic and
induced pluripotent stem cells. Cell Stem Cell 7,
84. Newman, A. M. & Cooper, J. B. Lab-specific gene
expression signatures in pluripotent stem cells. Cell
Stem Cell 7, 258–262 (2010).
85. Malchenko, S. et al. Cancer hallmarks in induced
pluripotent cells: new insights. J. Cell. Physiol. 225,
86. Ohm, J. E. et al. Cancer-related epigenome changes
associated with reprogramming to induced
pluripotent stem cells. Cancer Res. 70, 7662–7673
87. Hu, Q., Friedrich, A. M., Johnson, L. V. & Clegg, D. O.
Memory in induced pluripotent stem cells:
reprogrammed human retinal pigmented epithelial
cells show tendency for spontaneous redifferentiation.
Stem Cells 28, 1981–1991 (2010).
88. Ghosh, Z. et al. Persistent donor cell gene expression
among human induced pluripotent stem cells
contributes to differences with human embryonic stem
cells. PLoS ONE 5, e8975 (2010).
89. Ron-Bigger, S. et al. Aberrant epigenetic silencing of
tumor suppressor genes is reversed by direct
reprogramming. Stem Cells 28, 1349–1354 (2010).
90. Jirtle, R. L. Genomic imprinting and cancer. Exp. Cell
Res. 248, 18–24 (1999).
91. Lim, D. H. & Maher, E. R. Genomic imprinting
syndromes and cancer. Adv. Genet. 70, 145–175
276 | APRIL 2011 | VOLUME 11
© 2011 Macmillan Publishers Limited. All rights reserved
92. Stadtfeld, M. et al. Aberrant silencing of imprinted
genes on chromosome 12qF1 in mouse induced
pluripotent stem cells. Nature 465, 175–181 (2010).
93. Rugg-Gunn, P. J., Ferguson-Smith, A. C. & Pedersen,
R. A. Epigenetic status of human embryonic stem cells.
Nature Genet. 37, 585–587 (2005).
94. Rugg-Gunn, P. J., Ferguson-Smith, A. C. & Pedersen,
R. A. Status of genomic imprinting in human
embryonic stem cells as revealed by a large cohort of
independently derived and maintained lines. Hum.
Mol. Genet. 16, R243–R251 (2007).
95. Adewumi, O. et al. Characterization of human
embryonic stem cell lines by the International Stem
Cell Initiative. Nature Biotech. 25, 803–816 (2007).
96. Frost, J. M. et al. The effects of culture on genomic
imprinting profiles in human embryonic and fetal
mesenchymal stem cells. Epigenetics 6, 52–62 (2011).
97. Ghule, P. N. et al. Reprogramming the pluripotent cell
cycle: restoration of an abbreviated G1 phase in
human induced pluripotent stem (iPS) cells. J. Cell.
Physiol. 13 Oct 2010 (doi:10.1002/jcp.22440).
98. Singh, A. M. & Dalton, S. The cell cycle and Myc
intersect with mechanisms that regulate pluripotency
and reprogramming. Cell Stem Cell 5, 141–149 (2009).
99. Neganova, I., Zhang, X., Atkinson, S. & Lako, M.
Expression and functional analysis of G1 to S.
regulatory components reveals an important role for
CDK2 in cell cycle regulation in human embryonic
stem cells. Oncogene 28, 20–30 (2009).
100. Edel, M. J. et al. Rem2 GTPase maintains survival of
human embryonic stem cells as well as enhancing
reprogramming by regulating p53 and cyclin D1.
Genes Dev. 24, 561–573 (2010).
101. Holubcova, Z. et al. Human embryonic stem cells
suffer from centrosomal amplification. Stem Cells 29,
102. Okita, K., Ichisaka, T. & Yamanaka, S. Generation of
germline-competent induced pluripotent stem cells.
Nature 448, 313–317 (2007).
103. Wernig, M., Meissner, A., Cassady, J. P. & Jaenisch, R.
c-Myc is dispensable for direct reprogramming of
mouse fibroblasts. Cell Stem Cell 2, 10–12 (2008).
104. Mali, P. et al. Improved efficiency and pace of generating
induced pluripotent stem cells from human adult and
fetal fibroblasts. Stem Cells 26, 1998–2005 (2008).
105. Miura, K. et al. Variation in the safety of induced
pluripotent stem cell lines. Nature Biotech. 27,
First comparison of the tumorigenicity of mouse
iPSCs from different somatic origins.
106. Moriguchi, H., Chung, R. T. & Sato, C. An indicator for
evaluating the risk of cancerous transformations of
human induced pluripotent stem cells. Hepatology 51,
107. Gutierrez-Aranda, I. et al. Human induced pluripotent
stem cells develop teratoma more efficiently and faster
than human embryonic stem cells regardless the site
of injection. Stem Cells 28, 1568–1570 (2010).
Along with reference 86, this is one of the studies
directly comparing the tumorigenicity of HESCs and
108. Hentze, H., Graichen, R. & Colman, A. Cell therapy and
the safety of embryonic stem cell-derived grafts.
Trends Biotechnol. 25, 24–32 (2007).
109. Lee, A. S. et al. Effects of cell number on teratoma
formation by human embryonic stem cells. Cell Cycle
8, 2608–2612 (2009).
110. Hentze, H. et al. Teratoma formation by human
embryonic stem cells: evaluation of essential
parameters for future safety studies. Stem Cell Res. 2,
111. Eiges, R. et al. Establishment of human embryonic
stem cell-transfected clones carrying a marker for
undifferentiated cells. Curr. Biol. 11, 514–518 (2001).
112. Choo, A. B. et al. Selection against undifferentiated
human embryonic stem cells by a cytotoxic antibody
recognizing podocalyxin-like protein-1. Stem Cells 26,
113. Tan, H. L., Fong, W. J., Lee, E. H., Yap, M. & Choo, A.
mAb 84, a cytotoxic antibody that kills
undifferentiated human embryonic stem cells via
oncosis. Stem Cells 27, 1792–1801 (2009).
114. Fong, C. Y., Peh, G. S., Gauthaman, K. & Bongso, A.
Separation of SSEA-4 and TRA-1-60 labelled
undifferentiated human embryonic stem cells from a
heterogeneous cell population using magnetic-
activated cell sorting (MACS) and fluorescence-
activated cell sorting (FACS). Stem Cell Rev. 5, 72–80
115. Blum, B., Bar-Nur, O., Golan-Lev, T. & Benvenisty, N.
The anti-apoptotic gene survivin contributes to
teratoma formation by human embryonic stem cells.
Nature Biotech. 27, 281–287 (2009).
116. Smith, K. N., Singh, A. M. & Dalton, S. Myc
represses primitive endoderm differentiation in
pluripotent stem cells. Cell Stem Cell 7, 343–354
117. Varlakhanova, N. V. et al. myc maintains embryonic
stem cell pluripotency and self-renewal. Differentiation
80, 9–19 (2010).
118. Smith, K. & Dalton, S. Myc transcription factors:
key regulators behind establishment and maintenance
of pluripotency. Regen. Med. 5, 947–959 (2010).
119. Kim, J. et al. A Myc network accounts for similarities
between embryonic stem and cancer cell transcription
programs. Cell 143, 313–324 (2010).
120. Schuldiner, M., Itskovitz-Eldor, J. & Benvenisty, N.
Selective ablation of human embryonic stem cells
expressing a “suicide” gene. Stem Cells 21, 257–265
121. Krishna, K. A., Rao, G. V. & Rao, K. S. Stem cell-based
therapy for the treatment of Type 1 diabetes mellitus.
Regen. Med. 2, 171–177 (2007).
122. Korsgren, O. & Nilsson, B. Improving islet
transplantation: a road map for a widespread
application for the cure of persons with type I diabetes.
Curr. Opin. Organ Transplant. 14, 683–687 (2009).
123. Dean, S. K., Yulyana, Y., Williams, G., Sidhu, K. S. &
Tuch, B. E. Differentiation of encapsulated embryonic
stem cells after transplantation. Transplantation 82,
124. Giorgetti, A. et al. Generation of induced pluripotent
stem cells from human cord blood using OCT4 and
SOX2. Cell Stem Cell 5, 353–357 (2009).
125. Kim, J. B. et al. Direct reprogramming of human
neural stem cells by OCT4. Nature 461, 649–653
126. Russo, E. Follow the money—the politics of embryonic
stem cell research. PLoS Biol. 3, e234 (2005).
127. Hussein, S.M. et al. Copy number variation and
selection during reprogramming to pluripotency.
Nature 471, 58–62 (2011).
128. Gore, A. et al. Somatic coding mutations in human
induced pluripotent stem cells. Nature 471, 63–67
129. Lister, R. et al. Hotspots of aberrant epigenomic
reprogramming in human induced pluripotent stem
cells. Nature 471, 68–73 (2011).
Studies by the authors, described in this Review, have been
partially supported by funds from the Morasha-ISF (Grant No.
943/09) and Center of Excellence: The Legacy Heritage Fund
Program of The Israel Science Foundation (Grant No.
1801/10). The authors would like to thank E. Meshorer and
D. Ronen for critically reading the manuscript, and T. Golan-
Lev for her assistance with the figures.
Competing interests statement
The authors declare no competing financial interests.
NATURE REVIEWS | CANCER
VOLUME 11 | APRIL 2011 | 277
© 2011 Macmillan Publishers Limited. All rights reserved