© 2012 Nature America, Inc. All rights reserved.
Generation of induced pluripotent stem cells from
a small amount of human peripheral blood using a
combination of activated T cells and Sendai virus
718 | VOL.7 NO.4 | 2012 | nature protocols
Generating iPSCs is a prominent recent advance in stem cell
biology1. iPSCs have become cell sources for genetic disease
models and are expected to provide important new cell sources
for clinical therapies. Initial studies generated human iPSCs
from human fibroblasts obtained from dermal biopsy samples2,3.
However, although further studies successfully reprogrammed
several types of human somatic cells into iPSCs4–6, the methods
and cell sources most suitable for iPSC applications in humans
remain undetermined. In particular, the generation of iPSCs for
disease research should ideally avoid invasive tissue sampling,
which markedly reduces the number of patients who consent to
cell donations. In this regard, peripheral blood is an appealing cell
source because of the noninvasive collection and easy accessibility
of blood cells compared with skin fibroblasts and other types of
cells from adult tissues.
We recently demonstrated that transgene-free iPSCs can be effi-
ciently generated from a small amount of human peripheral blood
within 1 month of the blood sampling using a combination of
activated T cells and temperature-sensitive (TS) mutants of SeV
encoding human OCT3/4, SOX2, KLF4 and c-MYC (Fig. 1)7. We
named these T cell–derived iPSCs TiPS cells (TiPSCs). Recombinant
SeVs that replicate in the cytoplasm of infected cells in the form of
negative-sense single-stranded RNA were originally used to gen-
erate iPSCs from human fibroblasts8. These recombinant SeVs do
not integrate into the host genome9, and have already been used
in human iPSC generation using CD34 + cells from human cord
blood10. Introducing TS mutations also successfully erased residual
genomic RNA of the SeV vectors from the target cells10, thus gen-
erating transgene-free iPSCs with high efficiency. T cells are also
an appealing cell source because they are easily proliferated in vitro
using a plate-bound anti-CD3 monoclonal antibody and inter-
leukin (IL)-2 (ref. 11). Although it was reported that T cells are not
efficiently reprogrammed using only four factors in the mouse12,13,
SeV vectors were efficiently transduced into human activated T cells
to express exogenous genes14. Thus, the combination of activated
T cells and TS SeV mutants made it possible to generate TiPSCs
from patients effectively, easily and less invasively.
Advantages of the method
The initial methods for generating human iPSCs used a skin
biopsy2,3, requiring local anesthesia and suturation. In our pro-
tocol, iPSCs can be generated from patients without such inva-
sive tissue sampling. Sufficient patient-specific iPSCs can be
generated from 1 ml or less of peripheral blood, which contains
sufficient terminally differentiated T cells7. Our method might
therefore decrease the likelihood of patients refusing cell sam-
pling and therefore potentially increase the number of patients
who consent to generating iPSCs. In addition, TiPSCs can still
be generated from whole blood samples stored at room tem-
perature (20–25 °C) for 24 h and from mononuclear cells stored
at − 150 °C. Therefore, transported samples can be easily used
for generating iPSCs in any clinical situation. SeV also has the
possibility to be used for generating iPSCs from other human
blood cells such as monocytes, which, on the basis of existing
reports of iPSCs that were generated successfully using SeV from
T cells and CD34 + cells7,10, do not harbor T cell receptor (TCR)
or immunoglobulin gene rearrangements.
Comparison with other methods
In the first report of iPSC generation from human peripheral
blood cells6, mobilized CD34 + human peripheral blood cells were
successfully reprogrammed into iPSCs. However, this method
required extremely large amounts (~300 ml) of blood, an apher-
esis machine and drug administration before blood sampling
to mobilize the CD34 + blood cells, all of which should ideally
be avoided because of the possible associated side effects (e.g.,
bone pain), despite these effects being infrequent. Less invasive
methods using peripheral blood have also been reported for
Tomohisa Seki1, Shinsuke Yuasa1,2 & Keiichi Fukuda1
1Department of Cardiology, Keio University School of Medicine, Tokyo, Japan. 2Department of Cardiology, Center for Integrated Medical Research, Keio University School
of Medicine, Tokyo, Japan. Correspondence should be addressed to K.F. (firstname.lastname@example.org) or S.Y. (email@example.com).
Published online 15 March 2012; doi:10.1038/nprot.2012.015
Induced pluripotent stem cells (ipscs) have become important cell sources for genetic disease models, and they have the
potential to be cell sources for future clinical therapies. However, invasive tissue sampling reduces the number of candidates
who consent to donate cells for ipsc generation. In addition, integrated transgenes can potentially insert at inappropriate
points in the genome, and in turn have a direct oncogenic effect. technical modifications using a combination of activated
t cells and a temperature-sensitive mutant of sendai virus (seV) can avoid invasive tissue sampling and residual transgene
issues in generating ipscs. such advances may increase the number of consenting patients for cell donations. Here we present a
detailed protocol for the generation of ipscs from a small amount of human peripheral blood using a combination of activated
t cells and mutant seV encoding human oct3/4, soX2, KlF4 and c-MYc; t cell–derived ipscs can be generated within 1 month of
© 2012 Nature America, Inc. All rights reserved.
nature protocols | VOL.7 NO.4 | 2012 | 719
the successful reprogramming of mono-
nuclear blood cells15–17. In these methods,
mononuclear blood cells from donors or
frozen samples were infected using retro-
virus15 or lentivirus16,17 to express four
factors, human OCT3/4, SOX2, KLF4 and
c-MYC. In these studies, human T cell
reprogramming into iPSCs was achieved,
but the efficiency of reprogramming was
extremely low (approximately 0.0008–
0.01%). Although these methods used
less peripheral blood and did not require the pharmacological
pretreatment of patients, the problems of transgene genomic
insertion and low reprogramming efficiency remained, pre-
cluding their wide use in the clinical application of iPSCs.
Generating iPSCs with TS-mutated SeV easily erases residual
genomic viral RNA from the target cells10, and the method
is significantly more efficient (~0.1%) compared with those
protocols in which iPSCs were generated from T cells with
retrovirus or lentivirus.
Human keratinocytes derived from plucked human hair have also
been used as another less invasive method of obtaining iPSCs from
patient cells4,18. However, in some cases, these reported methods
require several hairs to obtain successful cell outgrowth of keratino-
cytes. Dental tissue has also been explored as a potential source of
iPSCs19. However, although teeth are routinely removed in many
clinics and no further procedures are required with respect to the
donor, it is generally difficult to routinely obtain patients’ dental
tissues—with specific genetic or nongenetic diseases—for the pur-
pose of iPSC studies. In comparison with these outlined methods,
our protocol involves harvesting only a small sample of peripheral
blood; in addition, T cell proliferation does not need stochastic cell
outgrowth. These are clear advantages for clinical application in
comparison with the methods reported in the past.
Blood sampling. Our protocol is focused on the simple procedure
of peripheral venous blood sampling to obtain the donor cells,
using a standard process. Patient somatic cells can then be easily
and aseptically obtained from the blood sample. In our protocol,
1 ml of whole blood is sufficient to generate TiPSCs (Fig. 2a).
Derivation of activated T cells. Peripheral blood mononuclear
cells (PBMCs) can be separated by a Ficoll gradient method from
heparinized whole blood samples (Fig. 2b). Although PBMCs
contain lymphocytes and monocytes, activation with plate-bound
anti-CD3 monoclonal antibody and IL-2 selectively proliferates
T cells, and clearly increases the proportion of T cells in the
cultured PBMCs11. CD3 protein exists in the complex of TCR
proteins on the surface of T cells, and can therefore be used
as a T cell–specific marker. Anti-CD3 antibody modulates the
TCR-CD3 complex to induce T cell proliferation and activation20,
whereas IL-2 also activates general T cell signaling pathways
and eventually promotes cytokine transcription, cell survival,
cell-cycle entry and growth21. At day 5 of culture with anti-CD3
monoclonal antibody and IL-2, CD3 + cells increased up to ~95%
of cultured PBMCs (Fig. 2c–e). With this culture method, users
can avoid using a fluorescence-activated cell sorter in which
Isolation of PBMCs
with cell count
IL-2 + anti-CD3 antibody +
ESC condition (bFGF +)
T cells before
on feeder cells
Day 5 Day 6Days 20–25
onto feeder cells
Figure 1 | Overview of the TiPSC generation
protocol. PBMCs are activated for 5 d with IL-2
and anti-CD3 antibody, and then transduced with
SeV expressing human OCT3/4, SOX2, KLF4 and
c-MYC. TiPSC colonies emerge at 20–25 d after
PBMCs day 0
500 µm500 µm
PBMCs day 5
PBMCs day 0
PBMCs day 5
Figure 2 | Isolation and activation of PBMCs. (a) Whole blood is collected into a 2.5-ml syringe by venipuncture. (b) Before
centrifugation, two distinct layers are distinguishable in the 15-ml tube. The upper, dark red layer is the diluted blood and the
lower layer is the Ficoll solution. After centrifugation, three distinct layers are apparent. The upper yellow layer contains platelet-
rich plasma, whereas the bottom clear layer is the Ficoll solution, and the thin white layer in between contains PBMCs (arrow).
(c) Morphology of PBMCs shortly after being seeded in the culture plate and activated using anti-CD3 antibody with IL-2 for 5 d.
Proliferated T cells show many clusters in the culture plate at day 5 of activation. (d,e) Flow cytometric analysis of isolated PBMCs
and PBMCs activated for 5 d with anti-CD3 antibody and IL-2 gated on the CD45 + cell population. CD3 surface expressions of these
populations were examined. The graph represents an average of three independent examinations. Error bars show means ± s.d.
© 2012 Nature America, Inc. All rights reserved.
720 | VOL.7 NO.4 | 2012 | nature protocols
the sorted cells are frequently damaged by laser emission and
the process of single-cell sorting.
Introduction of SeV vectors. SeV is an enveloped virus with a
single-stranded, negative-sense, nonsegmented RNA genome
belonging to the paramyxoviridae family. Recombinant SeV vec-
tors replicate only in the cytoplasm of infected cells9. SeV vectors
containing reprogramming factors were generated by introduc-
ing open reading frames for the human OCT3/4 (official symbol:
POU5F1) SOX2 and KLF4 genes into a fusion protein (F)-deficient,
TS SeV vector. SeV vector containing the c-MYC gene was also
generated with a more TS mutant, the TS15-SeV vector (P2,
L1361C, L1558I), so that it could be eliminated rapidly at 37 °C10.
The seed SeV/∆F vectors are generated by the transfecting template
pSeV/∆F carrying each transgene and pCAGGS plasmids—varying
the genes encoding T7 RNA polymerase, nucleoprotein (NP),
phosphoprotein (P), F5R and large protein (L)—into 293T cells.
Thereafter, the vector is propagated using LLC-MK2/F7/A cells,
which are SeV F–expressing LLC-MK2 cells10. SeV solutions can
be stored at − 80 °C and thawed before use. Activated PBMCs are
infected at day 5 of activation culture. For effective reprogramming
of T cells, this activation of PBMCs is important because it not only
increases the number of T cells, but also significantly promotes the
introduction efficiency of SeV. At a multiplicity of infection (MOI)
of 20, SeV could infect CD3 + T cells at >80% efficiency, which was
lower than 10% before activation (Fig. 3).
At around day 15 after transduction of SeV, human embryonic stem
cell (ESC)-like colonies emerge on the feeder cells (Fig. 4). These
T cell–derived TiPSC colonies show monoclonal TCR rearrange-
ment in their genome, which is a hallmark of mature terminally
differentiated T cells and indicates that each TiPSC colony is derived
from a single mature T cell.
PBMCs after activation
PBMCs before activation
Stored T cells
PBMCs 24 h left
Figure 3 | Infection of SeV. (a) Efficient induction of GFP by SeV in T cells activated for 5 d with anti-CD3 antibody and IL-2 at an
MOI of 20. Cells were viewed after a 24-h infection with SeV encoding GFP and another 24-h culture without SeV. (b) Flow cytometric
analysis of infected PBMCs gated on the CD3 + cell population. Activated and unactivated cells were analyzed after a 24-h infection
with SeV encoding GFP and another 24-h culture without SeV. (c) The graph shows an average of infection efficiencies of PBMCs
gated on the CD3 + cell population. Activated and unactivated cells were analyzed after a 24-h infection with SeV encoding GFP and another 24-h culture without
SeV. An average of three independent examinations are represented. Error bars show means ± s.d. (d) The graph shows an average of infection efficiencies of SeV
to activated PBMCs obtained from the frozen stocks of PBMCs, frozen stock of already activated PBMCs and whole blood stored for 24 h before use. These cells
were infected for 24 h with SeV encoding GFP at an MOI of 20. Analyses were done after another 24 h of culture without SeV. Flow cytometric analysis of the cells
gated on the CD3 + cell population in three independent examinations is shown. Error bars show means ± s.d.
50 µm50 µm
Figure 4 | Characterization of TiPSCs. (a) Example of a 10-cm dish stained
for ALP on day 20 after SeV infection at an MOI of 20 and seeded at a density
of 5 × 104 cells per 10-cm dish. Many ALP-positive T cell colonies that were
infected with SeV are visible. (b) ALP staining of TiPSC colonies. (c) Typical
ESC-like TiPS colony on day 20 after SeV infection. (d) Immunofluorescense
staining for pluripotency and surface markers (Nanog, Tra1-60, Oct 3/4,
SSEA3) in TiPSCs. Immunofluorescence staining was performed using the
following primary antibodies: anti-NANOG (RCAB0003P, ReproCELL), anti-
OCT3/4 (sc-5279, Santa Cruz), anti-SSEA3 (MAB4303, Millipore) and anti-
Tra1-60 (MAB4360, Millipore). DAPI (Molecular Probes) was used for nuclear
staining. The following secondary antibodies were used: anti-rabbit IgG
and anti-mouse IgG and IgM conjugated with Alexa Fluor 488 or Alexa Fluor
568 (Molecular Probes). (e) Hematoxylin and eosin–stained representative
teratomas derived from the TiPSC line.
© 2012 Nature America, Inc. All rights reserved.
nature protocols | VOL.7 NO.4 | 2012 | 721
Donors for blood sampling ! cautIon Subjects must have given informed
consent. ! cautIon All experiments involving humans must conform to
relevant governmental and institutional ethics regulations.
Novo-heparin (5,000 units per 5 ml; Mochida Pharmaceutical)
Ficoll-Paque PREMIUM (GE Healthcare, cat. no. 17-5442-02)
Purified NA/LE mouse anti-human CD3 (BD Pharmingen, cat. no.555336)
GT-T502 medium (Kohjin Bio, cat. no. 16025020)
Fetal bovine serum (FBS; Cell Culture Bioscience, cat. no. 171012)
Bovine albumin fraction V solution (BSA; Gibco, cat. no. 15260-037)
DMEM (Sigma, cat. no. D5546)
DMEM/F12 (Sigma, cat. no. D6421)
KnockOut serum replacement for ESCs/iPSCs (KSR; Gibco, cat. no. 10828-028)
GlutaMAX-I (Gibco, cat. no. 35050-061)
Non-essential amino acid solution (NEAA; Sigma, cat. no. M7145)
Penicillin-streptomycin (Gibco, cat. no. 15140-122)
2-Mercaptoethanol (2-ME; Invitrogen, cat. no. 21985-023)
! cautIon This solution is flammable, harmful if swallowed and toxic
when in contact with skin and eyes. Use protective gloves and safety
glasses when handling it.
Recombinant basic fibroblast growth factor, human (bFGF; Wako,
cat. no. 064-04541)
Collagenase type IV (Gibco, cat. no. 17104-019)
Gelatin powder (Sigma, cat. no. G-1890)
D-PBS( − ) (Wako, cat. no. 045-29795)
Acetamide (Wako, cat. no. 015-00115)
Propylene glycol (Wako, cat. no. 16-0499)
Cell Banker-2 (BIO LABO, cat. no. BLC-2)
TRIzol reagent (Invitrogen, cat. no. 15596-026)
Chloroform (Wako, cat. no. 038-02606)
Ethanol (Wako, cat. no. 057-00456)
Ethanol (70% (vol/vol); Wako, cat. no. 059-07895)
Isopropyl alcohol (Wako, cat. no. 166-04836)
SuperScript double-stranded cDNA synthesis kit (Invitrogen,
cat. no. 11917-010)
Oligo (dT)12–18 primer (Invitrogen, cat. no. 18418-012)
SYBR Premix Ex Taq II (Takara, cat. no. RR081A)
Sodium acetate (3 M; Wako, cat. no. 316-90081)
CytoTune-iPS reprogramming kit (OCT3/4-SeV/TS∆F, SOX2-SeV/TS∆F,
KLF4-SeV/TS∆F, c-MYC (HNL)-SeV/TS15∆F; Invitrogen,
cat. no. A13780-01)
Mitomycin C–treated mouse embryonic fibroblasts (MEFs; Reprocell,
cat. no. RCHEFC003)
MEF medium (see REAGENT SETUP)
Human iPSC medium (see REAGENT SETUP)
Isopropanol (Wako, cat. no. 16604836)
RNase-free water (Takara, cat. no. 9012)
DNase I (Invitrogen, cat. no. 18068-015)
Syringe (2.5 ml; Terumo, cat. no. SS-02LZ)
Butterfly needle (23 G; Terumo, cat. no. SV-23CLK)
Needle (23 G; Terumo, cat. no. NN-2325R)
Alcohol prep pads (Hakujuji)
Tourniquet (Asone, cat. no. MH-01)
Millex GV filter unit (0.22 µm; Millipore, cat. no. SLGV033RS)•
C-chip disposable hemocytometer (Digital Bio, cat. no. DHC-N01)
Trypan blue stain (0.4%; Gibco, cat. no. 15250)
Microtubes (1.5 ml; Thermo Fisher Scientific, cat. no. 131-615C)
Tube (50 ml; Corning, cat. no. 430829)
Tube (15 ml; Corning, cat. no. 430053)
Tissue culture dish (10 cm; Falcon, cat. no. 353003)
Tissue culture plate (96 well; Falcon, cat. no. 353078)
Tissue culture plate (24 well; Falcon, cat. no. 353047)
Tissue culture plate (12 well; Falcon, cat. no. 353043)
Tissue culture plate (6 well; Falcon, cat. no. 353046)
Cryovials (1.5 ml; Sumilom, cat. no. MS-4702X)
Freezing container (Sanyo, cat. no. MDM-U73V)
Cell culture incubator set at 37 °C, 5% CO2 (Sanyo, cat. no. MCO-18AIC)
Versatile refrigerated centrifuge (Sanyo, cat. no. AX-320)
NanoDrop 2000 (Thermo Fisher Scientific, cat. no. ND-2000)
ABI 7500 real-time PCR system (Applied Biosystems, cat. no. 7500-01)
bFGF Prepare 0.1% (wt/vol) BSA/PBS in a sterile tube and use it to dissolve
bFGF for a final concentration of 4 ng ml − 1. Prepare 100-µl aliquots in screw-
cap microcentrifuge tubes and store them at − 20 °C.
Gelatin-coated culture dishes Dissolve 1 g of gelatin powder in 1,000 ml
of distilled water, autoclave, filter the solution with a 0.22-µm Millex GV
filter unit and store it at 4 °C. Add an appropriate volume of 0.1% (wt/vol)
gelatin solution to cover the entire area of the culture dishes to coat. Incubate
the dishes for at least 30 min at 37 °C in a sterile environment. Remove the
gelatin solution before use.
Collagenase type IV solution Dissolve 1 g of collagenase type IV
powder in 1,000 ml of DMEM/F12 medium and filter the solution with
a 0.22-µm Millex GV filter unit. Make 50-ml aliquots in 50-ml tubes and
store them at − 20 °C. Thawed solution can be stored at 4 °C for up to
1 week before use.
Anti-CD3 monoclonal antibody–coated plates Dissolve anti-human CD3
antibody in D-PBS( − ) to a concentration of 10 µg ml − 1. Add the anti-human
CD3 antibody solution to 24-well tissue culture plates to soak the surface
of each plate, and then incubate them at 37 °C in a 5% CO2 incubator for at
least 30 min. Remove the anti-human CD3 antibody solution and wash the
plates once with D-PBS( − ) before seeding the cells.
Human iPSC medium To prepare 500 ml of human iPSC medium, mix
387.5 ml of DMEM/F12 medium with 100 ml of KSR, 5 ml of GlutaMAX-I
(1 mM), 5 ml of penicillin-streptomycin, 5 ml of NEAA (10 µM), 500 µl of
2-ME (100 µM) and 50 µl of bFGF (4 ng ml − 1). Filter the medium with a
0.22-µm filter unit and store it for up to 1 week at 4 °C.
MEF medium To prepare 500 ml of MEF medium, mix 450 ml of
DMEM medium with 50 ml of FBS and 2.5 ml of penicillin-streptomycin.
Filter the medium with a 0.22-µm filter unit and store it for up to 2 weeks
at 4 °C.
DAP213 solution To prepare 10 ml of DAP213 solution, mix 5.37 ml of
human iPSC medium, 1.43 ml DMSO, 1 ml of 10 M acetamide and 2.2 ml of
propylene glycol. Store the solution for up to 1 month at − 80 °C.
SeV solutions To prepare working stocks of SeV solutions (from the Cyto-
Tune kit), thaw the solutions on ice and prepare 50-µl stocks in 1.5-ml tubes.
Working stocks can be stored at − 80 °C.
Blood sampling ● tIMInG ~10 min
1| Sterilize the cap from the bottle of heparin with an alcohol prep pad.
! cautIon Wash your hands before starting venipuncture. Wear gloves when handling blood. Change gloves after
venipuncture in each donor or if the gloves become contaminated.
© 2012 Nature America, Inc. All rights reserved.
722 | VOL.7 NO.4 | 2012 | nature protocols
2| Combine a 23-G needle and 2.5-ml syringe and draw up 100–300 µl of heparin.
3| Release the 23-G needle from heparinized 2.5-ml syringe and combine a new 23-G butterfly needle with the heparinized
4| Identify the median cubital or cephalic veins of donors’ arms; then palpate and trace the vein paths with the
5| Sterilize the venipuncture site with an alcohol prep pad.
! cautIon Do not palpate the venipuncture site after sterilization.
6| Apply the tourniquet above the selected puncture site.
! cautIon Do not place the tourniquet too tightly or leave it on for more than 3 min.
7| Remove the needle shield and perform venipuncture. Insert the needle into the blood vessel and hold still once a
backflow of blood is seen in the tube of the butterfly needle setup.
! cautIon Venipuncture must be done by a person who is well trained and legally certified to carry out the procedure.
! cautIon Patients’ informed consent must be obtained before blood sampling.
8| Draw 1–2 ml of blood into the syringe.
9| Remove the tourniquet.
! cautIon Do not withdraw the needle before removing the tourniquet.
10| Withdraw the needle fully, apply pressure to the alcohol prep pad over the puncture site and maintain the pressure for
3–5 min until the bleeding stops.
11| Discard the needle of the Vacutainer into a biohazard container without recapping the needle.
! cautIon Dispose of items that are used for venipuncture immediately and in appropriate containers.
crItIcal step This step and all subsequent steps should be carried out using sterile reagents and equipment.
pause poInt Heparinized whole blood can be stored for up to 24 h before use (Fig. 2a).
Isolating pBMcs using Ficoll gradient ● tIMInG ~1 h
12| Add 1–2 ml of heparinized whole blood to a 15-ml tube.
13| Add 1–2 ml of D-PBS( − ) and dilute the blood with D-PBS( − ) 1:1.
14| Prepare 3 ml of Ficoll-Paque PREMIUM in a separate 15-ml tube.
15| Pour the diluted blood onto the Ficoll solution carefully so as to form two layers.
! cautIon Do not mix the blood and the Ficoll solution. The blood must remain on top.
16| Centrifuge for 30 min at 400g at room temperature.
17| Collect the layer of PBMCs without touching the Ficoll layer, using sterile pipette tips, into a new 15-ml tube (Fig. 2b).
crItIcal step When collecting PBMCs, carefully avoid obtaining any of the Ficoll solution layer. Contamination of the
Ficoll solution reduces the collection rate of the mononuclear cells.
18| Dilute the PBMCs by adding 5 ml of D-PBS( − ).
19| Centrifuge for 5 min at 200g at room temperature.
20| Discard the supernatant, add 5 ml of D-PBS( − ) and dilute the PBMCs with D-PBS( − ).
21| Centrifuge for 5 min at 200g at room temperature.