Genetic correction and analysis of induced pluripotent
stem cells from a patient with gyrate atrophy
Sara E. Howdena,b,c, Athurva Gored, Zhe Lid, Ho-Lim Fungd, Benjamin S. Nislere, Jeff Niea, Goukai Chena,b,c,
Brian E. McIntosha,b,c, Daniel R. Gulbransona,b,c, Nicole R. Diola,b,c, Seth M. Taapkene, David T. Vereidea,b,c,
Karen Dyer Montgomerye, Kun Zhangd, David M. Gammf, and James A. Thomsona,b,c,g,1
aMorgridge Institute for Research, Madison, WI 53715;bDepartment of Cell and Regenerative Biology, University of Wisconsin School of Medicine and Public
Health, Madison, WI 53707-7365;cGenome Center of Wisconsin, University of Wisconsin, Madison, WI 53706;dDepartment of Bioengineering, University of
California at San Diego, La Jolla, CA 92093;eWicell Research Institute, Madison, WI 53707;fDepartment of Ophthalmology and Visual Sciences, Waisman
Center Stem Cell Research Program, Eye Research Institute, University of Wisconsin, Madison, WI 53705; andgDepartment of Molecular, Cellular, and
Developmental Biology, University of California, Santa Barbara, CA 93106
Contributed by James A. Thomson, March 1, 2011 (sent for review January 21, 2011)
Gene-corrected patient-specific induced pluripotent stem (iPS) cells
offer a unique approach to gene therapy. Here, we begin to assess
whether the mutational load acquired during gene correction of
iPS cells is compatible with use in the treatment of genetic causes
of retinal degenerative disease. We isolated iPS cells free of trans-
gene sequences from a patient with gyrate atrophy caused by a
point mutation in the gene encoding ornithine-δ-aminotransferase
(OAT) and used homologous recombination to correct the genetic
defect. Cytogenetic analysis, array comparative genomic hybridiza-
tion (aCGH), and exome sequencing were performed to assess the
genomic integrity of an iPS cell line after three sequential clonal
events: initial reprogramming, gene targeting, and subsequent
removal of a selection cassette. No abnormalities were detected
after standard G-band metaphase analysis. However, aCGH and
exome sequencing identified two deletions, one amplification,
and nine mutations in protein coding regions in the initial iPS cell
clone. Except for the targeted correction of the single nucleotide in
the OAT locus and a single synonymous base-pair change, no ad-
ditional mutations or copy number variation were identified in iPS
cells after the two subsequent clonal events. These findings con-
firm that iPS cells themselves may carry a significant mutational
load at initial isolation, but that the clonal events and prolonged
cultured required for correction of a genetic defect can be accom-
plished without a substantial increase in mutational burden.
acquired and inherited diseases, but the extensive culture period
required could introduce a mutational burden incompatible with
specific clinical applications. For example, diabetes is a disease
that continues to have a serious impact on quality and length
of life despite its clinical management with insulin and is an
attractive target for iPS cell-based transplantation therapies.
However, pancreatic cancer is generally fatal within a few months
of diagnosis. Thus, it will be essential to determine the accu-
mulated mutational load of cells before transplantation and to
determine whether these mutations may lead to a significantly
increased risk of oncogenesis. A recent study revealed that a
significant number of point mutations are generally present in
iPS cell clones relative to the average sequence of the parental
somatic cells (4). Exome sequencing of 22 independent human
iPS cell lines reprogrammed by using four different methods
revealed a range of 1–14 point mutations in each of the lines
analyzed, with a projected average of six protein-coding muta-
tions per genome.
Because of the proliferative potential of iPS cells, homologous
recombination can be used to correct specific genetic defects
prolonged culture, drug selection, and additional clonal genetic
bottlenecks beyond initial iPS cell generation and may, therefore,
be expected to introduce high mutational loads. Here, we assess
the mutational load accumulated in iPS cells after three clonal
utologous therapies based on induced pluripotent stem
(iPS) cells (1–3) have the potential to treat a wide range of
events that led to the targeted correction of a single base-pair
mutation in the OAT locus.
Homozygous mutations in the OAT locus result in Gyrate at-
rophy (GA) of the choroid and retina, a disease characterized by
progressive loss of visual acuity and night vision with eventual loss
of central vision typically occurring in the fourth to fifth decades
of life (5). Although the exact pathophysiological mechanism of
GA remains unknown, the retinal pigmented epithelium (RPE) is
thought to be the initial site of degeneration (6). RPE dysfunction
is also involved in macular degeneration, the leading cause of
incurable blindness in developed countries, affecting >14 million
people worldwide (7). Diseases involving the RPE are attractive
targets for iPS cell-based therapies, because there are already
protocols for deriving RPE from human iPS cells (8, 9) and the
site can be easily monitored with noninvasive techniques. Fur-
thermore, thephysiological functionoftheRPEdepends onfairly
simple anatomical relationships so it is reasonable to expect that
transplanted cells could sustain retinal function if the structural
damage is not yet too severe. Indeed, a recent study has dem-
onstrated protective effects of iPS-derived RPE after trans-
plantation into RCS dystrophic rats (10), and the two most recent
human embryonic stem (ES) cell clinical trials approved in the
United States involves transplantation of RPE cells (11).
In this study, we used an episomal reprogramming method
(12) to generate iPS cells free of transgene sequences from a GA
patient homozygous for a deleterious mutation in the OAT gene.
We then used homologous recombination with a BAC-based
vector to correct the disease-causing mutation. Vector-free,
correctly targeted iPS cell clones were identified by custom high-
density array comparative genomic hybridization (aCGH) and
confirmed by standard techniques. We also performed sequential
aCGH and exome sequencing of the parental fibroblast line, the
iPS cell line before targeting, the iPS cell line after gene tar-
geting, and the iPS cell line after removal of the selection cas-
sette to assess the accumulation of genomic changes during these
procedures. We found that although there was a fairly substantial
mutational load in the iPS cell line at the time of derivation
before targeting, homologous recombination and cassette re-
moval could be carried out with minimal additional changes.
These results suggest that the accumulation of mutations may
Author contributions: S.E.H., K.Z., D.M.G., and J.A.T. designed research; S.E.H., A.G., Z.L.,
H.-L.F., B.S.N., J.N., G.C., B.E.M., D.R.G., N.R.D., S.M.T., and K.D.M. performed research;
A.G., Z.L., H.-L.F., B.S.N., J.N., S.M.T., and K.D.M. contributed new reagents/analytic tools;
S.E.H., A.G., Z.L., H.-L.F., B.S.N., J.N., S.M.T., D.T.V., K.D.M., and K.Z. analyzed data; and
S.E.H. and J.A.T. wrote the paper.
The authors declare no conflict of interest.
Freely available online through the PNAS open access option.
1To whom correspondence should be addressed. E-mail: jthomson@morgridgeinstitute.
This article contains supporting information online at www.pnas.org/lookup/suppl/doi:10.
| April 19, 2011
| vol. 108
| no. 16
not limit the clinical utility of gene targeting approaches, but that
extensive genetic characterization of the initial iPS cell clones
will be critical.
We obtained primary dermal fibroblasts from a GA patient from
the Coriell Institute for Medical Research. This donor is homo-
zygous for a C > T transition at nucleotide 677 (677C > T) in exon
7 of the OAT gene that converts alanine 226 (GCG) to valine
(GTG). To generate patient-specific iPS cells free of transgene
sequences, we used an episomal reprogramming strategy based
on the transfection of oriP-containing plasmid vectors encoding
seven reprogramming factors (OCT4, SOX2, NANOG, LIN28,
c-Myc, KLF4, and SV40 large T-antigen) (12). These EBV-based
episomal vectors are replicated in synchrony with the host cells
and are relatively stable, but in the absence of selection, are
nonetheless lost from dividing cells at a rate of ≈2–6% per cell
generation (13, 14), allowing the vectors to persist long enough
to induce reprogramming before being ultimately lost, resulting
in iPS cells that are free of transgene sequences. After a sin-
gle transfection, iPS cell clones were selected based on basic
morphological criteria (e.g., compact colonies, high nucleus-to-
cytoplasm ratios, and prominent nucleoli) and six clones were
expanded for further analysis. Total DNA was extracted and PCR
analysis was performed to examine whether episomal DNA used
for reprogramming remained. In four iPS cell clones, episomal
sequences were not detected (Fig. S1).
Two iPS cell lines (5 and 12) were selected for subsequent
gene targeting experiments. We designed a gene targeting con-
struct carrying a 37.4-kb fragment from a human BAC clone that
encodes the entire OAT coding region. A puromycin resistance
cassette flanked by loxP sites was inserted just downstream of the
OAT coding region, resulting in two homologous arms ≈27.7 kb
and 8.8 kb in size (Fig. 1). We successfully isolated and expanded
15 and 5 puromycin-resistant colonies after the electroporation
of the linearized targeting vector into iPS cell clones 5 and 12,
respectively. Genomic DNA was extracted from each clone, and
PCR analysis was performed to test for the presence of vector
backbone sequences because clones that had undergone the
desired homologous recombination event should not contain
these sequences (Fig. S2). To assess expression of wild-type (w/t)
OAT mRNA, total RNA was extracted and RT-PCR was per-
formed to amplify exon 7 of the OAT transcript, which was then
sequenced. Exon 7 of the OAT gene was also amplified from the
genomic DNA of each iPS cell clone and sequenced. After
scoring each line for the presence or absence of vector backbone
sequences and a w/t copy of the OAT gene and/or transcript
(Table S1), we identified two potential gene-targeted iPS cell
clones (5.12 and 12.4). These two clones also displayed an almost
1:1 ratio of mutant to w/t transcript as evidenced by sequencing
analysis of the OAT RT-PCR products (Fig. 2A). Fluorescent in
situ hybridization (FISH) analysis revealed only two signals
corresponding to the endogenous OAT loci on chromosome
10q26 and failed to detect any additional random integration
events in these clones (Fig. 2B). FISH analysis performed on an
iPS cell clone in which backbone targeting vector sequences were
detected by PCR revealed the two endogenous loci and a third
signal corresponding to a random integration event (Fig. 2C).
Although we confirmed a gene targeting event in 2 of 20 (10%)
drug-resistant clones, we could not detect a w/t copy of the OAT
gene or transcript or any vector backbone sequences in 10 (50%)
clones. Therefore, we cannot exclude that these clones did un-
dergo homologous recombination but with a crossing over event
occurring between exon 7 and the puromycin cassette rather than
upstream of exon 7 (Fig. 1).
We next wished to examine whether aCGH could also be used
as a method to distinguish gene-targeted iPS cell clones from
drug-resistant clones arising from random integration of the
gene targeting vector. Genomic DNA from a gene-targeted iPS
line (clone 12.4) and a random integrant (clone 5.15) confirmed
by FISH were mixed with sex-mismatched reference DNA and
then hybridized to a custom-designed array containing probes
spanning our OAT gene targeting vector. No copy number var-
iation (CNV) involving the OAT coding sequence could be
detected for the gene-targeted line (clone 12.4) or its parental
iPS cell line, whereas increased hybridization corresponding to
the integrated puromycin cassette was observed as expected (Fig.
3A). Furthermore, CNV was not detected across the endogenous
OAT locus, providing further evidence that homologous re-
combination of the gene targeting vector had occurred in iPS cell
line 12.4 (Fig. 3B). In contrast, in the random integrant (5.15),
we detected more than two copies of the OAT gene along with
the puromycin cassette and all of the gene targeting vector
backbone sequences (Fig. 3 A and B).
Ourcustom arraywasalso designed toinclude probesequences
spanning the three episomal vectors used for reprogramming. We
detected no vector sequences for the gene-targeted line (clone
12.4) and its parental iPS cell line (clone 12), but an ≈1,850-bp
region of the reprogramming vectors corresponding to the 5′ end
of oriP and C-terminal portion of the EBNA1 gene was detected
in both the random integrant (clone 5.15) and its parental iPS cell
line (clone 5)(Fig.2C).PCR analysis using primersspecific tothis
sequence confirmed its presence in iPS cell lines 5, 5.12, and 5.15
to confirming homologous recombination, aCGH is useful in
excluding any iPS cell clones with integrated vector sequences.
Although all clones had normal karyotypes after G-banding
analysis, aCGH analysis revealed a number of subkaryotypic
alterations in both the parental and gene-corrected iPS cells that
were not detected after aCGH analysis of the original patient
fibroblast line. These genomic alterations included two small
deletions in the random integrant (clone 5.15) and its parental
line (clone 5), and two separate small deletions and a single
amplification in the gene-targeted line (clone 12.4) and its pa-
rental line (clone 12) (Table 1). No additional genomic aberra-
tions were detected by aCGH in the gene-targeted iPS cell line
(clone 12.4) compared with its parental iPS cell line (clone 12).
Analysis of iPS cells derived from the same patient line but using
lentiviral vectors also revealed several other amplifications and de-
letions, indicating that reprogramming with episomal vectors per
se is unlikely to be the cause of large-scale genomic aberrations.
To remove the puromycin cassette, we transfected in vitro
transcribed mRNA-encoding Cre recombinase into the gene-
targeted iPS cell line (clone 12.4). Loss of the selection cassette
was observed in 8/8 independent colonies as evidenced by sen-
sitivity to puromycin in the culture medium and by PCR analysis
of the extracted genomic DNA (Fig. S4A). One clone (12.4.2)
was selected for further analysis. Southern blot confirmed the
loss of the puromycin cassette and G-band analysis confirmed
a normal karyotype (Fig. S4 B and C). Furthermore, when cells
from this line were injected into immunocompromised mice, we
the study. A loxP-flanked selection cassette (PGK-Puro) was inserted ≈2 kb
downstream of the OAT coding region by recombineering. The sizes of the
homologous arms are shown. The point mutation in exon 7 (E7*), the
regions amplified by PCR to identify random integrants, and the probe used
in Southern blot experiments are also indicated.
Schematic diagram of the BAC-based gene targeting vector used in
| www.pnas.org/cgi/doi/10.1073/pnas.1103388108Howden et al.
observed teratomas comprising all three primary germ layers
(Fig. S5). This line was also analyzed by aCGH, which also
revealed loss of the puromycin cassette, lack of CNV at the OAT
locus, absence of any reprogramming vector sequences, and lack
of additional changes in genome integrity in comparison with its
parental iPS cell lines (Fig. S6).
To investigate the accumulation of mutations through reprog-
ramming, prolonged iPS cell culture, and the serial clonal events
needed for gene targeting, we performed exome sequencing on
the parental patient fibroblast cell line, an uncorrected iPS cell
line (clone 12), the gene-targeted iPS cell line (clone 12.4), and
the cassette-free iPS cell line (clone 12.4.2). We searched for
novel mutations between each successive clonal line relative to
its direct progenitor cell line sequence and discovered nine
mutations in the original iPS cell clone (Table 2). Somewhat
surprisingly, the genomes were remarkably stable through the
subsequent clonal events. Exome sequencing successfully detec-
ted the correct 677T > C substitution introduced by homologous
recombination in the gene-corrected iPS cell lines (clones 12.4
and 12.4.2). Although all nine original mutations were confirmed,
noadditional mutations weredetected in thetargeted iPScell line
(clone 12.4), and only one additional synonymous mutation was
detected in the cassette-free iPS cell line (clone 12.4.2).
A central challenge to the therapeutic use of either human iPS
or ES cells is understanding what specific types and number of
mutations and genomic aberrations are clinically acceptable for
a given application, for such changes will always accumulate with
continued cell division. Indeed, recent analyses of large sets of
iPS cell lines identified a substantial number of cell lines carrying
full and partial chromosomal aberrations that arose either from
culture adaptation or from clonal selection of the parental cell
line (15, 16). Although all of the iPS cell lines in this study had
normal karyotypes by standard G-band analysis, aCGH and
exome sequencing revealed at least three regions of CNV and
nine mutations in the original iPS cell line before targeting.
These findings are comparable to the results of other recent
studies, suggesting that iPS cells themselves may carry a signifi-
cant number of genomic aberrations and protein-coding muta-
tions arising either in the donor, during somatic cell expansion,
or during reprogramming itself (4). Importantly, we found that
our current culture conditions can allow extensive expansion and
genetic manipulations of human iPS without a dramatic increase
in genomic instability and mutational burden. Clearly additional
studies are required to better estimate the mutational load ac-
quired during iPS cell culture, yet our results nonetheless allow
an initial crude estimate of this load. The observation that only
a single point mutation was acquired after gene correction and
cassette removal is consistent with the expected load based on
the average mutation rate in human somatic cells, while also
being above the expected load based on average mutation rate in
human germ-line cells (17).
Although BAC vectors have widely been used for the genetic
modification of murine ES cells, only a single study had reported
the use of BAC-based vectors for gene targeting in human plu-
ripotent stem cells (18). Here, we demonstrate the use of a BAC-
based homologous recombination strategy to successfully repair
a single base-pair mutation in a patient-specific iPS cell line that
is also free of reprogramming transgenes. We also found that
transfection of mRNA-encoding Cre recombinase is a highly
efficient method for removal of a loxP-flanked selection cassette
in gene-corrected iPS cells. Although the residual 35-bp loxP site
that remains in the genome of the cassette-free iPS cells is un-
likely to have any detrimental effect, it would nonetheless be
desirable to have the correction be entirely seamless. A seamless
correction could be achieved by performing two sequential ho-
mologous recombinations, using both positive and negative se-
lection, or by flanking the selection cassette with terminal repeats
Patient iPSCs Gene corrected iPSCs:
Gene corrected iPSCs:
targeting vector or cDNA from an uncorrected iPS cell line and gene-corrected iPS cell lines arising from a random integration or gene targeting event. Arrows
indicate the location of the point mutation in the patient line. The antisense strand is shown. FISH analysis of a gene-targeted iPS cell line (B) and a random
integrant (C) using a fluorescently labeled probe specific to the OAT locus. Yellow arrows indicate endogenous signals on chromosome 10q; red arrow
indicates a third signal caused by random integration of the gene-targeting construct.
Gene correction of patient iPS cells. (A) Chromatograms resulting from Sanger sequencing of OAT exon 7 PCR products amplified from the gene
Howden et al. PNAS
| April 19, 2011
| vol. 108
| no. 16
from the piggyBac transposon, which can be seamlessly excised
after transient expression of the piggyBac transposase (19).
In this study, we used an episomal approach to produce
integration-free iPS cells, which has the advantage of requiring
just a single transfection. Although there are now several other
approaches to produce vector-free iPS cells, including seamless
removal of piggyBac vectors (20), protein transduction (21), and
RNA transfection (22), previous exome sequencing studies failed
to find differences in mutational load between iPS cells derived
by integration-free episomal or RNA methods, or between
methods that included SV40 large T-antigen and those that did
not (4). Although episomal sequences have the potential to in-
tegrate into the genome, as evidenced in this study, the detailed
characterization required to assess genomic integrity of iPS cell
clones could simultaneously be used to identify and exclude any
iPS cell clones carrying integrated transgenes with no additional
cost to the analysis. We demonstrate the feasibility of using
customized aCGH for this purpose.
The mutational load observed in our iPS cell line is within the
range of mutations detected in the recent exome sequencing
study of 22 different iPS cell lines (4). Interestingly, analyses in
that report revealed that a majority, but not all, of these muta-
tions were present in subpopulations of the parental cell lines.
Therefore, simply changing iPS cell generation methods or cul-
ture conditions cannot eliminate such mutations. Because iPS
cell lines from different patients will also have unique sets of
mutations, it will be extremely challenging to assign relative risks
to different combinations of mutations. Clearly an ongoing di-
reprog. vector 1
iPS 12.4 (targeted)
iPS 5.15 (random)
three reprogramming vectors (C) are shown for gene-corrected iPS cells resulting from a gene targeting (clone 12.4) or random integration (clone 5.15) event.
Data for the two parental iPS cells (clones 12 and 5) are also shown.
High-density aCGH analysis of gene-corrected iPS cells. aCGH data for the OAT gene targeting vector (A), endogenous OAT locus (B), and one of the
Table 1.Genomic aberrations identified in patient iPS cell lines by aCGH
iPS line ChromsomeRegion SizeType Genes
12, 12.4 1p21.3
5.15 1 (OAT)
*Patient iPS line generated with lentiviral reprogramming vectors.
| www.pnas.org/cgi/doi/10.1073/pnas.1103388108Howden et al.
alogue between stem cell and cancer biologists will be needed as
the rapid advances in genomic sequencing yield greater insights
into the genesis of cancer. Further work is also needed to de-
termine whether there is a consistently different mutational load
in different starting somatic cell types influencing the choice of
cellular source for iPS cell generation, and whether the muta-
tional load increases dramatically with aging, which might justify
the banking of neonatal cells. With the advent of high-density
arrays and modern sequencing methods, even minor changes in
genome integrity can be detected, allowing an extensive evalu-
ation of genome integrity and mutational load. The mutational
burden observed in iPS cells and its impact on clinical utility are
critical issues that must be addressed before the broad thera-
peutic application of gene-corrected iPS cells in transplantation
Materials and Methods
iPS Cell Derivation and Culture. Patient fibroblasts were obtained from Coriell
Laboratories (ID no. GM06330) and cultured in DMEM (Invitrogen) supple-
mented with 10% FBS (HyClone) at 37 °C, 5% CO2, 5% O2. The episomal
plasmids and protocols used for reprogramming were described (12). After
isolation, iPS cells were maintained and expanded in TeSR and passaged
routinely with EDTA as described (23).
Transfection. Gene targeting experiments were performed by using con-
ditions optimized for human ES cells (24). Cells were harvested with TrpLE
(Invitrogen) and resuspended at a concentration of 5 × 106cells per mL;
0.5 mL of the cell suspension was mixed with ≈50 μg of linearized gene
targeting vector in 300 μL of PBS and electroporated in a 0.4-cm cuvettete.
Cells were then plated on a 10-cm matrigel-coated dish in TeSR medium
with 10 μM HA100 (Sigma). Puromycin was added to the culture medium
(1 ng/mL) after 3 d. To remove the loxP-flanked selection, cassette cells were
harvested and electroporated as described with 15 μg in vitro transcribed
mRNA-encoding Cre recombinase.
FISH. Metaphase spreads were prepared according to a standard procedure.
The OAT gene targeting vector, labeled with Spectrum Green-dUTP (Vysis),
was used as the probe.
Genomic PCR and RT-PCR. Total genomic DNA was extracted by using the
Nucleon BACC3 kit (GE Healthcare). Total RNA was extracted by using the
RNeasy Plus Mini Kit (Qiagen). cDNA was synthesized by using SuperScript III
(Invitrogen), according to the manufacturer’s protocol. PCR was performed
by using the Taq PCR Core Kit (Qiagen).
aCGH. A custom array was designed that consisted of the NimbleGen human
CGH 2.1M Whole-Genome-Tiling v2.0D array in addition to 4,633 custom
designed probes that cover the gene targeting construct and the reprog-
ramming plasmid vectors used in this study. The original whole-genome-
tilling design has the resolution of 5 kb over the entire human genome,
whereas our custom fine-tilling design could detect 200-bp CNV regions. CGH
experiments were performed at Roche NimbleGen and the WiCell Cytoge-
netics Laboratory (WiCell Research Institute). Experiments performed at
WiCell involved a multistep labeling procedure as described (25). Data were
analyzed using SignalMap software (Roche NimbleGen). aCHG data can be
found at http://www.ncbi.nlm.nih.gov/geo/ under accession no. GSE26773.
Whole-Genome Library Construction. Whole-genome libraries for each cell line
were constructed as described (4). Briefly, genomic DNA for each of the four
samples was sheared by using a Covaris AFA and enzymatically end-repaired.
Sheared DNA was then enzymatically ligated to common sequencing pri-
mers and amplified by using PCR to generate a whole-genome library.
In-Solution Hybridization Capture with DNA Baits. Whole-genome libraries
were enriched for exomic regions by using the NimbleGen SeqCap EZ Exome
kit. Libraries were denatured and hybridized to the SeqCap EZ Exome library
and then captured using streptavidin beads by following the manufacturer’s
protocol. Enriched libraries were sequenced on an Illumina Genome
Consensus Sequence Generation and Variant Calling. Novel reprogramming-
associated mutations were identified as described (26). Briefly, Illumina GA IIx
reads were filtered by using GERALD and mapped to the genome using
Bowtie or BWA. Consensus sequence generation was performed by using
SAMtools or GATK. Each iPS consensus sequence was directly compared with
its progenitor consensus sequence, and candidate sites indicating a gain of
a new allele in the iPS line were chosen.
Sanger Validation of Candidate Mutations. Each candidate mutation was
validated via amplification with specially designed primers and capillary
Sanger sequencing performed by GeneWiz.
ACKNOWLEDGMENTS. We thank Dr. David Frisch and Dr. Fred Blattner for
providing the hES-2A–inducible BAC vector. This study is funded by a Wynn-
Gund Translational Award from the Foundation Fighting Blindness and par-
tially funded by the Charlotte Geyer Foundation and National Institutes of
Health Grant R01 HL094963. S.E.H. is supported by a National Health and
Medical Research Council Overseas Biomedical Fellowship. A.G. is supported
by the Focht-Powell Fellowship and a California Institute for Regenerative
Medicine predoctoral Fellowship.
1. Takahashi K, et al. (2007) Induction of pluripotent stem cells from adult human
fibroblasts by defined factors. Cell 131:861–872.
2. Takahashi K, Yamanaka S (2006) Induction of pluripotent stem cells from mouse
embryonic and adult fibroblast cultures by defined factors. Cell 126:663–676.
3. Yu J, et al. (2007) Induced pluripotent stem cell lines derived from human somatic
cells. Science 318:1917–1920.
4. Gore A, et al. (2011) Somatic coding mutations in human induced pluripotent stem
cells. Nature 471:63–67.
Table 2. Mutations in protein coding regions identified via exome sequencing
iPS cell line
position, bases)GeneCodons AA Sub.SNP type
Gene mutated in cancer?
12, 12.4, 12.4.215,40166820,G/A
N/A (not applicable) is listed for the three mutations that are synonymous or the gene-corrected base-pair change that was brought about by homologous
recombination. Synonymous changes do not alter amino acid sequence and therefore should not be damaging to the cell.
Howden et al.PNAS
| April 19, 2011
| vol. 108
| no. 16
5. Simell O, Takki K (1973) Raised plasma-ornithine and gyrate atrophy of the choroid Download full-text
and retina. Lancet 1:1031–1033.
6. Wang T, Milam AH, Steel G, Valle D (1996) A mouse model of gyrate atrophy of the
choroid and retina. Early retinal pigment epithelium damage and progressive retinal
degeneration. J Clin Invest 97:2753–2762.
7. Gehrs KM, Anderson DH, Johnson LV, Hageman GS (2006) Age-related macular
degeneration—emerging pathogenetic and therapeutic concepts. Ann Med 38:
8. Buchholz DE, et al. (2009) Derivation of functional retinal pigmented epithelium from
induced pluripotent stem cells. Stem Cells 27:2427–2434.
9. Meyer JS, et al. (2009) Modeling early retinal development with human embryonic
and induced pluripotent stem cells. Proc Natl Acad Sci USA 106:16698–16703.
10. Carr AJ, et al. (2009) Protective effects of human iPS-derived retinal pigment
epithelium cell transplantation in the retinal dystrophic rat. PLoS ONE 4:e8152.
11. Anonymous (2011) Stem-cell trial. Nature 469:137.
12. Yu J, et al. (2009) Human induced pluripotent stem cells free of vector and transgene
sequences. Science 324:797–801.
13. Kirchmaier AL, Sugden B (1995) Plasmid maintenance of derivatives of oriP of Epstein-
Barr virus. J Virol 69:1280–1283.
14. Sugden B, Warren N (1988) Plasmid origin of replication of Epstein-Barr virus, oriP,
does not limit replication in cis. Mol Biol Med 5:85–94.
15. Laurent LC, et al. (2011) 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.
16. Mayshar Y, et al. (2010) Identification and classification of chromosomal aberrations
in human induced pluripotent stem cells. Cell Stem Cell 7:521–531.
17. Lynch M (2010) Rate, molecular spectrum, and consequences of human mutation.
Proc Natl Acad Sci USA 107:961–968.
18. Song H, Chung SK, Xu Y (2010) Modeling disease in human ESCs using an efficient
BAC-based homologous recombination system. Cell Stem Cell 6:80–89.
19. Fraser MJ, Ciszczon T, Elick T, Bauser C (1996) Precise excision of TTAA-specific
lepidopteran transposons piggyBac (IFP2) and tagalong (TFP3) from the baculovirus
genome in cell lines from two species of Lepidoptera. Insect Mol Biol 5:141–151.
20. Woltjen K, et al. (2009) piggyBac transposition reprograms fibroblasts to induced
pluripotent stem cells. Nature 458:766–770.
21. Zhou H, et al. (2009) Generation of induced pluripotent stem cells using recombinant
proteins. Cell Stem Cell 4:381–384.
22. Warren L, et al. (2010) Highly efficient reprogramming to pluripotency and directed
differentiation of human cells with synthetic modified mRNA. Cell Stem Cell 7:618–630.
23. Chen G, Hou Z, Gulbranson DR, Thomson JA (2010) Actin-myosin contractility is
responsible for the reduced viability of dissociated human embryonic stem cells. Cell
Stem Cell 7:240–248.
24. Zwaka TP, Thomson JA (2003) Homologous recombination in human embryonic stem
cells. Nat Biotechnol 21:319–321.
25. Taapken SM, Nisler BS, Sampsell-Barron TL, Dyer Montgomery K (2010) Automation
in an array comparative genomic hybridization laboratory improves throughput and
data quality. JALA 15:390–395.
26. Quail MA, Swerdlow H, Turner DJ (2009) Improved protocols for the illumina genome
analyzer sequencing system. Curr Protoc Hum Genet 62:18.2.1–18.2.27.
| www.pnas.org/cgi/doi/10.1073/pnas.1103388108 Howden et al.