EZH1 iPSC CAR T

Article
EZH1 repression generates mature iPSC-derived

CAR T cells with enhanced antitumor activity
Graphical abstract Authors
Ran Jing, Irene Scarfo,
Mohamad Ali Najia, ..., Trista E. North,
Marcela V. Maus, George Q. Daley
Correspondence
[email protected]
In brief
Jing et al. combine a stroma-free
differentiation system with EZH1-
repression-mediated epigenetic
reprogramming to generate
developmentally mature iPSC-derived
CAR T cells with enhanced antitumor
activities.
Highlights
d Stroma-free differentiation of iPSCs into T cells expressing
diverse TCRs
d EZH1 repression generates mature EZ-T cells similar to pe-

ripheral blood ab T cells
d EZ-T cells can give rise to memory-like T cells upon activation
d CAR EZ-T cells display enhanced antitumor activity in vitro

and in vivo
Jing et al., 2022, Cell Stem Cell 29, 1181–1196

August 4, 2022 ª 2022 Elsevier Inc.
https://doi.org/10.1016/j.stem.2022.06.014 ll
ll
Article
EZH1 repression generates mature iPSC-derived
CAR T cells with enhanced antitumor activity
Ran Jing,1,2 Irene Scarfo,3,4 Mohamad Ali Najia,1,2,5,6 Edroaldo Lummertz da Rocha,7 Areum Han,1,2 Michael Sanborn,1
Trevor Bingham,1 Caroline Kubaczka,1,2 Deepak K. Jha,1,2 Marcelo Falchetti,8 Thorsten M. Schlaeger,1,9
Trista E. North,1,9,10 Marcela V. Maus,3,4 and George Q. Daley1,2,9,10,11,*
1Stem Cell Program, Boston Children’s Hospital, Boston, MA 02115, USA
2Department of Biological Chemistry and Molecular Pharmacology, Harvard Medical School, Boston, MA 02115, USA
3Cellular Immunotherapy Program, Massachusetts General Hospital Cancer Center, Charlestown, MA 02114, USA
4Department of Medicine, Harvard Medical School, Boston, MA 02115, USA
5Broad Institute of MIT and Harvard, Cambridge, MA 02142, USA
6Harvard-MIT Health Sciences & Technology, Massachusetts Institute of Technology, Cambridge, MA 02142, USA
7Department of Microbiology, Immunology and Parasitology, Federal University of Santa Catarina, Florianópolis, SC 88040-900, Brazil
8Graduate Program of Pharmacology, Center for Biological Sciences, Federal University of Santa Catarina, Florianópolis, SC 88040-900,
Brazil
9Division of Hematology/Oncology, Boston Children’s Hospital and Dana Farber Cancer Institute, Boston, MA 02115, USA
10Developmental and Regenerative Biology Program, Harvard Medical School, Boston, MA 02115, USA
11Lead contact
*Correspondence: [email protected]
https://doi.org/10.1016/j.stem.2022.06.014
SUMMARY
Human induced pluripotent stem cells (iPSCs) provide a potentially unlimited resource for cell therapies, but
the derivation of mature cell types remains challenging. The histone methyltransferase EZH1 is a negative
regulator of lymphoid potential during embryonic hematopoiesis. Here, we demonstrate that EZH1 repres-
sion facilitates in vitro differentiation and maturation of T cells from iPSCs. Coupling a stroma-free T cell dif-
ferentiation system with EZH1-knockdown-mediated epigenetic reprogramming, we generated iPSC-
derived T cells, termed EZ-T cells, which display a highly diverse T cell receptor (TCR) repertoire and mature
molecular signatures similar to those of TCRab T cells from peripheral blood. Upon activation, EZ-T cells give
rise to effector and memory T cell subsets. When transduced with chimeric antigen receptors (CARs), EZ-T
cells exhibit potent antitumor activities in vitro and in xenograft models. Epigenetic remodeling via EZH1
repression allows efficient production of developmentally mature T cells from iPSCs for applications in adop-
tive cell therapy.
INTRODUCTION bor-intensive, and expensive therapy is still not widely available.

An alternative and more readily accessible source for CAR T cells
Chimeric antigen receptor (CAR) T cell-based cancer immuno- is needed.
therapy has proven remarkably effective against lymphoid malig- Human induced pluripotent stem cells (iPSCs) represent an
nancies (June et al., 2018). In this approach, T cells collected ideal source for the scalable manufacture of off-the-shelf prod-
from a patient’s blood are expanded in vitro and engineered to ucts for cell therapy. An early study explored the possibility of us-
express tumor antigen-specific CARs so that the resulting CAR ing human iPSCs to generate T cells for adoptive T cell therapy
T cells are capable of recognizing and attacking tumor cells. and showed that iPSC-derived T cells engineered with a CAR
CAR T cell therapy has demonstrated durable therapeutic effi- against the CD19 antigen were capable of inhibiting tumor
cacy and holds great promise for the cure of lymphoid malig- growth in tissue culture and murine models (Themeli et al.,
nancies, but the broader application of this breakthrough anti- 2013). However, these iPSC-CAR T cells displayed the transcrip-
cancer strategy has been impeded by several factors. To date, tional signature of innate-like gd T cells and were not as function-
patients have been their own donors, and those who have ally robust as mature ab T cells. More recent efforts have demon-
been heavily pre-treated sometimes lack adequate numbers of strated enhanced T cell maturation when employing iPSCs
functional T cells available for autologous harvest (Fesnak generated from T cell sources that carry productively rearranged
et al., 2016a, 2016b). Additionally, the expansion and engineer- T cell receptors (TCRs) (Iriguchi et al., 2021) or incorporate orga-
ing of autologous T cells requires time that some patients cannot noid or thymic culture systems (Vizcardo et al., 2018; Montel-Ha-
afford. Consequently, this highly effective yet cumbersome, la- gen et al., 2019). Therefore, methods that enable efficient
Cell Stem Cell 29, 1181–1196, August 4, 2022 ª 2022 Elsevier Inc. 1181
ll
Article
A B
D E F
(legend on next page)
1182 Cell Stem Cell 29, 1181–1196, August 4, 2022

ll
Article
stroma-free production of fully functional, developmentally sion of endogenous or transgenic TCR has been shown to
mature iPSC-T cells are needed to realize a broad range of produce T cells with innate-like features (Terrence et al., 2000;
iPSC-based adoptive T cell therapies. Baldwin et al., 2005; Egawa et al., 2008; Zhao et al., 2007).
Recent studies have revealed key roles for epigenetic regula- Though it has been shown that co-culture with OP9-DL4 cells
tors during definitive hematopoiesis and lymphoid development. yields iPSC-T cells with a broad TCR repertoire (Chang et al.,
Our lab identified EZH1, a component of polycomb repressive 2014), whether a stroma-free system can fully recapitulate
complex 2 (PRC2), as a critical negative regulator of definitive normal T cell development and support developmental matura-
lymphoid commitment during embryonic hematopoietic devel- tion and random TCR rearrangement in vitro remains unclear.
opment (Vo et al., 2018). In this study, we tested the hypothesis To answer this question, we sought to produce highly differenti-
that EZH1 repression might lead to more faithful in vitro recapit- ated T cells from non-T cell-derived iPSCs while avoiding animal
ulation of T cell differentiation and developmental maturation stromal cells and serum. The differentiation procedure is illus-
from iPSCs. Below, we show that iPSC-T cells derived in a trated in Figure 1A. In the first step, human erythroblast-derived
stroma-free, serum-free system following EZH1 knockdown iPSCs (cell line 1157) were induced to form embryoid bodies to
(EZ-T cells) display a molecular signature that more closely ap- generate hemogenic endothelial (HE) cells with hematopoietic
proximates peripheral blood ab T cells than innate-like gd potential using a previously reported protocol (Ditadi et al.,
T cells and that upon activation are capable of giving rise to 2015). The CD34+ HE cells were collected and seeded as a mono-
effector cytotoxic T cells and T cell subsets that exhibit a mem- layer on tissue culture dishes coated with Notch Delta-like Ligand
ory-like phenotype. EZ-T cells expressing anti-CD19 CARs 4 (DLL4) and VCAM-1 (Shukla et al., 2017). To initiate T cell spec-
showed more robust tumor-killing activity and cytokine secretion ification, HE cells were cultured in SFEMII media supplemented
compared with control CAR-loaded iPSC-T cells generated with a cocktail of cytokines (SCF, FLT3, IL-7, IL-3, TPO). During
without EZH1 repression. Injection of CAR-loaded EZ-T cells this time period, HE cells underwent endothelial-to-hematopoiet-
into B cell lymphoma-bearing mice led to superior persistence ic transition (EHT), giving rise to floating cells that contain
and more efficient tumor clearance than control iPSC-derived CD5+CD7+ T cell progenitors (proTs). On day 14, the floating he-
CAR T cells. Thus, a combination of EZH1 repression and matopoietic cells were collected and replated on new DLL4
stroma-free T cell differentiation allows efficient production of coated plates, followed by 3 weeks of culture with FLT3 and
mature iPSC-T cells with enhanced functionality. Such an IL-7. The proT cells continued to expand and passed through a
approach is compatible with commercial-scale production for transient CD3-CD4+ immature CD4 single positive (ISP) stage
CAR T cell-based immunotherapy. before activating the expression of CD8 to form CD4+CD8+ dou-
ble positive (DP) cells (Figures 1B and 1C). After 5 weeks of differ-
RESULTS entiation, CD3+TCRab+ T cells were detected. Further culture of
these cells in the presence of anti-CD3/CD28 antibodies and IL-
A serum-free, stroma-free system allows efficient 15 facilitated the induction of single positive (SP) cells. At week 6,
differentiation of iPSCs into T cells expressing a majority of cells expressed CD3, with the population consisting
diverse TCRs of both DP and CD4/CD8 SP T cells (Figure 1D).
Existing in vitro T cell differentiation protocols largely rely on en- To further evaluate the yield and efficiency of our stroma-free
gineered mouse stromal cells expressing Notch ligands, such protocol, we differentiated iPSCs into T cells using both the
as OP9-DL1/DL4 or MS5-DL1 cells, to provide the continuous stroma-free method and a previously published OP9-DL1 co-
Notch signaling needed for T cell development (Holmes and Zú- culture system (Themeli et al., 2013). The stroma-free protocol
€cker, 2009; Timmermans et al., 2009; Seet et al., 2017).
ñiga-Pflu resulted in a significant increase in CD3+ T cell production (Fig-
Previous studies have shown that immobilized Notch ligands ure 1E). Compared with OP9-DL1 co-culture, the stroma-free
support the induction of primary CD34+ hematopoietic stem or system also led to more efficient T cell-specific differentiation,
progenitor cells (HSPCs) into T cell lineages (Huijskens et al., indicated by an increased proportion of T cells and concomitant
2014; Shukla et al., 2017). Recently, a stroma-free differentiation reductions of non-lymphoid cell frequencies (Figures 1F and
protocol was reported to generate T cells from TCR-transduced S1A). To determine whether the stroma-free differentiation was
iPSCs or antigen-specific cytotoxic T cell (CTL)-derived iPSCs accompanied by normal TCR rearrangement, we extracted
(Iriguchi et al., 2021). These iPSCs carry pre-rearranged TCRs genomic DNA from stroma-free iPSC-T cells and sequenced
and display different T cell differentiation kinetics than their the complementarity-determining region 3 (CDR3) of the TCRb
wild-type counterparts (Themeli et al., 2013). Premature expres- locus to profile the TCR repertoire. ImmunoSEQ analysis
Figure 1. Stroma-free differentiation of human iPSCs into T cells

(A) Schematic illustration of the stroma-free T cell differentiation.
(B) Representative images showing day-0 CD34+ HE cells, day-14 proT cells, and day-35 T cells. Scale bars, 200 mm.
(C) Expression of T cell lineage-specific markers during the stroma-free T cell differentiation of iPSCs (gated on CD45+ cells).
(D) Frequencies of CD4, CD8, and DP T cells in CD3+ cells after 6 weeks of differentiation (n = 3, mean ± SEM).
(E) Numbers of week-6 CD3+ T cells generated via OP9-DL1 (blue) or stroma-free (red) differentiation, normalized to numbers of CD34+ HE cells seeded on day
0 (n = 3, mean ± SEM, * p % 0.05).
(F) Frequencies of B cells (CD19+), NK cells (CD56+), myeloid cells (CD33+), T cell precursors (CD5+CD3-), and T cells (CD3+) in CD45+ hematopoietic cells after
6 weeks of T cell differentiation using the OP9-DL1 (blue) or stroma-free (red) method (n = 3, mean ± SEM, * p % 0.05).
(G) Frequencies of Vb family genes determined by sequencing of the TCRb CDR3 region (n = 2).
Cell Stem Cell 29, 1181–1196, August 4, 2022 1183

ll
Article
identified tens of thousands of unique rearrangements with a phenotypic and molecular analyses to characterize the nature of
high degree of diversity in the usage of Vb family genes, demon- T cells produced following EZH1 knockdown. It has been sug-
strating the random recombination of CDR3 in the iPSC-T cells gested that T cell-derived iPSCs (T-iPSCs) bearing rearranged
(Figure 1G). Taken together, these data establish a stroma-free endogenous TCR genes result in the premature expression of
system that faithfully recapitulates normal T cell development TCR receptors during in vitro differentiation (Themeli et al.,
to produce iPSC-T cells with a highly diverse TCR repertoire. 2013), whereas T cell differentiation from non-T cell-derived-
PSCs displays similar kinetics to cord blood (CB) CD34+ cells,
EZH1 repression facilitates in vitro T cell differentiation with both gradually upregulating surface TCR/CD3 expression
from iPSCs only after the DP stage (Seet et al., 2017) (Figure 1C).
Previous studies have shown that EZH1 acts as a key regulator After differentiation, EZ-T cells displayed a significant increase
of hematopoietic multipotency, and repression of EZH1 function in CD3+TCRab+ and decrease in CD3+TCRgd+ T cells
promotes lymphoid potential during mouse and zebrafish em- (Figures 2F, 2G, S2E, and S2F), indicating that EZH1 knockdown
bryonic development (Vo et al., 2018; Soto et al., 2021). To deter- promotes differentiation toward ab T cell fate rather than gd
mine whether the inhibition of EZH1 would facilitate in vitro T cell T cells. EZH1 knockdown also resulted in a decrease in CD1a,
differentiation from iPSCs, we performed shRNA-mediated which is expressed in thymocytes but not mature peripheral
EZH1 knockdown during T cell specification from iPSC-derived blood T cells, further demonstrating that EZ-T cells exhibit a
CD34+ HE cells (Figures S1B and S1C). EZH1-knockdown HE more mature T cell phenotype (Figures 2H and S2G). Unlike
cells were then induced to differentiate into EZ-T cells via the mature, conventional cytotoxic T cells that express the CD8ab
stroma-free system and compared with control iPSC-SF-T heterodimer, innate-like gd T cells or intestinal intraepithelial lym-
cells derived as above (Figure 2A). EZH1 knockdown produced phocytes (IELs) express CD8 as a CD8aa homodimer, which has
a 2-fold increase in live cells after 2 weeks of differentiation been considered a less robust co-receptor and may even sup-
(Figure 2B) and after 6 weeks resulted in both a higher proportion press TCR activation (Bosselut et al., 2000; Holler and Kranz,
of T cells among CD45+ hematopoietic cells and an increased 2003; Cheroutre and Lambolez, 2008). Studies using OP9-DL1
absolute number of T cells generated from each starting stroma have yielded CD8aa T cells (Themeli et al., 2013; Maeda
CD34+ cell, suggesting enhanced T cell differentiation specificity et al., 2016), while recent protocols using 3D artificial thymic or-
and efficiency (Figures 2C, 2D, and S1D). Similar results were ganoids or a Notch ligand-based feeder-free system have
obtained in multiple iPSC lines that were derived from distinct yielded iPSC-T cells that express CD8ab (Montel-Hagen et al.,
donors using different reprogramming strategies, indicating 2019; Iriguchi et al., 2021). Similarly, our stroma-free differentia-
that the EHZ1-knockdown-mediated enhancement of T cell dif- tion protocol supports the production of CD8ab T cells, with
ferentiation was not cell line restricted (Figure S1E). As an alter- EZH1 knockdown promoting higher yields of CD8ab T cells
native to the shRNA-mediated EZH1 knockdown, we engineered and barely detectable quantities of innate-like CD8aa T cells
a doxycycline-inducible CRISPR interference (CRISPRi) (Figures 2I and S2H). Collectively, these analyses revealed that
construct into iPSC-derived CD34+ HE cells to transcriptionally repression of EZH1 promotes efficient in vitro differentiation of
repress EZH1 expression (Figure S2A). Interestingly, CRISPRi- iPSC cells into mature SP T cells.
mediated EZH1 knockdown during T cell specification (weeks
0–2) led to a significant increase in the production of CD3+ EZ-T cells display a molecular signature similar to that
T cells, while induction of EZH1-CRISPRi after the formation of of peripheral blood TCRab T cells
proT cells (weeks 2–5) failed to promote T cell differentiation (Fig- To assess the extent to which our iPSC-T cells resemble their
ure S2B). In control cells, the expression of EZH1 significantly in vivo counterparts, we evaluated the fidelity of stroma-free
increased during specification of HE into proT cells and was in vitro T cell differentiation via CellNet, a network biology plat-
downregulated after the proT stage (Figure S2C). EZH2, a homo- form used to analyze the faithfulness of cell fate conversions
log of EZH1 that likewise acts as a catalytic subunit of the PRC2 and the similarity between in vitro-derived cell types and gold-
complex, was highly expressed at the later stages of T cell differ- standard native-tissue equivalents (Cahan et al., 2014; Radley
entiation in both control and EZH1 shRNA-treated cells (Fig- et al., 2017). Analysis of RNA-seq gene expression profiles by
ure S2D). Such an observation is in agreement with previous CellNet revealed a high degree of similarity between iPSC-
findings that EZH1, but not EZH2, functions in HE cells to inhibit derived T cells and donor-derived T cells isolated from peripheral
lymphoid lineage commitment (Vo et al., 2018) and explains why blood mononuclear cells (PBMC), and a clear discrimination
the inhibition of EZH1 had no impact on later stages of T cell dif- from less-differentiated multipotent HSPCs (Figures 3A and
ferentiation. After 6 weeks of differentiation, the control iPSC-SF- S3A). We next sought a more refined analysis and therefore per-
T cells contained a substantial proportion of DP T cells, whereas formed RNA-seq to compare the gene expression profile of EZ-T
the EZ-T cells consisted predominantly of CD8 or CD4 SP cells cells with PBMC-derived TCRab T cells, PBMC-TCRgd T cells,
(Figure 2E). In contrast to the limited production of SP T cells and PBMC-NK cells, as well as T cells generated from CB
when differentiated on OP9-DL1 stroma, (La Motte-Mohs CD34+ HSPCs or iPSCs in the absence of EZH1 knockdown
et al., 2005; Montel-Hagen and Crooks, 2019), our data demon- via in vitro differentiation on OP9-DL1 stroma or our stroma-
strate that a combination of stroma-free differentiation with free system. Examining the expression of T cell signature genes,
EZH1-knockdown supports more efficient induction of SP cells. we found that iPSC-T cells differentiated via our stroma-free pro-
As previous studies have shown that iPSC-derived T cells tend tocol without EZH1 knockdown (PSC-SF-T) were more similar to
to resemble innate-like gd T cells (Nishimura et al., 2013; Viz- CB-HSPC-derived T cells than were iPSC-T cells differentiated
cardo et al., 2013; Themeli et al., 2013), we performed immuno- via OP9-DL1 stroma (Figure 3B). In contrast, the EZ-T cells

ll
Article
A B
C D E
F G
H I
Figure 2. EZH1 knockdown facilitates in vitro T cell differentiation from iPSCs

(A) Schematic illustration of EZ-T cell generation.
(B) Numbers of live cells during T cell differentiation using control (blue) or EZH1 knockdown (KD) (red) iPSC-derived HE cells (n = 3, mean ± SEM,
* p % 0.05, ** p % 0.01).
(C) Frequencies of CD3+ T cells in CD45+ cells generated from control (blue) or EZH1 KD (red) cells after stroma-free T cell differentiation (n = 3, mean ± SEM,
*** p % 0.001).
(D) Numbers of CD3+ T cells generated from control (blue) or EZH1 KD (red) cells, normalized to numbers of CD34+ HE cells seeded on day 0 (n = 3, mean ± SEM,
** p % 0.01).
(E) Frequencies of CD4, CD8, and DP T cells generated from control (blue) or EZH1 KD (red) cells after 6 weeks of stroma-free T cell differentiation (n = 3, mean ±
SEM, *** p % 0.001).
(F–H) Expression of CD3 and TCRab /TCRgd/CD1a in control and EZH1 KD cells after stroma-free T cell differentiation, gated on CD45+ cells.
(I) Expression of CD8a and CD8b in control and EZH1 KD cells after stroma-free T cell differentiation, gated on CD8 T cells.
were most similar to PBMC TCRab T cells. Furthermore, we that was most similar to ab T cells (Figure 3C). To investigate the
compared the transcriptional signature of PBMC-derived TCRab molecular mechanism underlying the mature phenotype of EZ-T
T with that of TCRgd T cells and identified a list of genes that can cells, we performed gene set enrichment analysis (GSEA) on the
be used to distinguish ab T from gd T cells. Expression levels of most significantly upregulated genes in EZ-T cells compared
these genes were determined across all cell types, and the result with control iPSC-SF-T cells (lacking EZH1 knockdown), and
again showed that EZ-T cells exhibited a gene expression profile observed that these genes were highly enriched in biological

ll
Article
A B C
D E
Figure 3. EZ-T cells display molecular features of mature TCRab T cells

(A) Heatmap showing CellNet analysis of RNA-seq data from iPSC-derived CD34+ HSPCs and iPSC-derived T cells generated via stroma-free protocol, either
with EZH1 KD (iPSC-EZ-T) or without (iPSC-SF-T).
(B) Dendrogram representing hierarchical cluster analysis based on expression of TCR pathway genes (BIOCARTA_TCR_PATHWAY, M19784) (n = 3).
(C) Heatmap showing unsupervised clustering analysis based on TCRab signature genes (n = 3).
(D) Top GO terms of biological processes enriched in iPSC-EZ-T cells versus iPSC-SF-T cells by GSEA analysis (n = 3).
(E) GSEA enrichment plots showing over-representation of gene sets related to T cell development and functions.
processes directly associated with T cell differentiation or func- pressed TRAC, TRBC2, CD2, and CD7, and showed a greater
tion (Figures 3D and 3E). To access whether the differences iden- downregulation of residue ‘‘innate’’ genes (TRDC, KLRB1) (Fig-
tified by bulk RNA-seq were due to changes in cell type compo- ure S3B). Collectively, EZH1 repression during in vitro iPSC dif-
sitions, we performed single-cell RNA sequencing (scRNA-seq) ferentiation promotes the production of ab T cells that display
analysis on control iPSC-SF-T cells and EZ-T cells and a more mature phenotype.
compared gene expression profiles for the sub-populations of We next performed immunosequencing to determine the TCR
TCRab T cells (TRAC+TRDC-). A relatively small number of genes repertoire of EZ-T cells and observed a high degree of TCRb
were differentially expressed between control iPSC-SF ab T cells diversity with no preferential Vb gene usage (Figure 4A), suggest-
and those with EZH1 knockdown (Table S1). Interestingly, ing that EZH1-knockdown did not cause significant
TCRab T cells with EZH1 knockdown more abundantly ex- clonal expansion of individual T cells with specific TCR

ll
Article
B C
Figure 4. TCR repertoire analysis of EZ-T cells

(A) Frequencies of Vb family genes in EZ-T cells determined by sequencing of the TCRb CDR3 region (n = 2).
(B) TCRb CDR3 length of iPSC-EZ-T cells (red) compared with control iPSC-SF-T cells without EZH1 KD (blue) (n = 2).
(C) Relative expression levels of TdT/DNTT in iPSC-HE cells (week 0) or iPSC-T cells with (red) or without (blue) EZH1 KD (week 4, n = 3, mean ± SEM, * p % 0.05).
rearrangements. Notably, EZ-T cells displayed longer CDR3 re- from iPSCs have previously been shown to display a naive
gions than iPSC-SF-T cells without EZH1 knockdown (Fig- phenotype (Seet et al., 2017; Iriguchi et al., 2021), but which
ure 4B). iPSC-derived T cells tend to have shorter CDR3 regions T cell subsets can be derived from naive iPSC-T cells is still
than mature peripheral blood T cells, likely due to lower expres- largely unknown. To answer this question, we used scRNA-seq
sion levels of terminal deoxynucleotidyl transferase (TdT), the to profile the distinct changes and characterize the T cell subsets
enzyme responsible for random nucleotide insertion during that arise during T cell differentiation and activation. Upon the
VDJ recombination, which is encoded by the DNTT gene (Mon- completion of EZ-T cell differentiation, we sorted CD45+ he-
tel-Hagen et al., 2019). To test this, we determined the expres- matopoietic cells before and after T cell activation and generated
sion levels of TdT/DNTT in both iPSC-derived HE cells and 11,131 single-cell transcriptomes. We identified eight clusters
week-4 DP T cells harboring control or EZH1 shRNA. As ex- across all samples, with the majority of cells manifesting a
pected, TdT/DNTT was only expressed in DP T cells and more T cell fate (CD5+) (Figures 5A and 5B). Consistent with previous
abundantly expressed in cells with EZH1-knockdown (Fig- immunophenotypic analyses, cells were predominantly CD8
ure 4C). These data suggest that EZH1-knockdown during SP cells, with lower numbers of CD4 SP and DP T cells (Fig-
iPSC differentiation promotes T cell maturation, with the result- ure 5B). Further analyses on the signature gene expression
ing EZ-T cells exhibiting molecular signatures that most closely profiles revealed only small proportions of innate NK-like
resemble peripheral blood TCRab T cells. (CD56+CD5-KLRB1+) or gd T cells (TRDC+TRGC1+). For
the mature T cell compartment, we identified two naive-like
EZ-T cell subsets display effector and memory-like T cell clusters (CCR7+SELL+IL2RAlowLEF1high) distinguished
phenotypes by their cycling status (Willinger et al., 2006), two effector-
After exiting the thymus, newly produced naive T cells give rise to like clusters (CCR7-SELL-GZMBhighGZMAhighNKG7+), and
effector and memory subsets, which are characterized by notably, a memory-like T cell cluster that expresses low levels
distinct phenotypic and functional features and cumulatively of cytotoxicity-related genes (GZMB, NKG7) and highly
shape T cell immunity (Kumar et al., 2018). T cells generated expresses genes that have been linked to memory T cell

ll
Article
A B
F G
Figure 5. Single-cell RNA-seq analysis identifies memory-like T cell subsets in EZ-T cells after activation
(A) Uniform manifold approximation and projection (UMAP) visualization of all the CD45+ cells generated from EZ-T cell differentiation, with and without activation.
(B) UMAP visualization of the expression of hematopoietic and T cell markers.
(C) Heatmap showing expression levels of T/NK cell signature genes across all clusters.
(D) GSEA analysis of the memory-like CD8 cluster showing over-representation of genes enriched in memory T cells but not naive or effector T cells.
(E) UMAP analysis comparing cell types in CD45+ cells generated from EZ-T cell differentiation, before and after activation.
(F) Proportion of cells in unactivated and activated CD45+ cells generated via EZ-T cell differentiation.
(G) CellRouter analysis showing transcriptional regulators enriched in the memory-like CD8 T cell cluster.

ll
Article
identities (CCR7+SELL+IL7R+CD2highCCL5highFAShighEOMEShigh) et al., 2019) (Figure S5A). Almost all the CD45+ cells derived
(Sanders et al., 1988; Huster et al., 2004; Intlekofer et al., 2005; via EZ-T cell differentiation overlapped with peripheral blood
Marçais et al., 2006). We further validated the presence of a T cells; cells that represent early HSPCs or other non-lymphoid
memory-like T cell subset by detecting the expression of lineages were barely detectable (Figure S5B). Consistent with
CD45RA, CD45RO, and CCR7 (Figure S4A). Moreover, it is previous analysis, following activation, a subset of EZ-T cells
known that cytotoxic T cells can express NK cell genes, emerged as CD8 central memory T cells (Figure S5B). A cross
including inhibitory NK cell receptors that raise the threshold of comparison between iPSC-derived T cells with their in vivo coun-
TCR stimulation and dampen T cell responses (Vivier and An- terparts also indicates that the memory-like CD8 cluster in EZ-T
fossi, 2004). Among the inhibitory NK cell receptors, KLRB1 is cells closely correlates with peripheral blood central memory
expressed by tumor infiltrating effector T cells for several types CD8 T cells (Figure S5C). Taken together, iPSC-derived EZ-T
of human cancer, negatively regulating their antitumor activity cells recapitulate the differentiation of naive T cells that give
(Mathewson et al., 2021). A more recent study further identified rise to effector cells and T cell subsets that exhibit a memory-
KLRB1, together with other NK cell receptors, as a hallmark of like phenotype.
exhausted, dysfunctional CAR T cells (Good et al., 2021). In
our cell populations, the memory-like T cell cluster expressed CAR T cells generated from EZ-T cells exhibit enhanced
lower levels of NK cell genes, including KLRB1, which further dis- antitumor activity
tinguishes these cells from the terminally differentiated effector- Having shown that EZ-T cells display molecular features similar
like T cell clusters (Figure 5C). In addition to the expression of to those of mature peripheral blood TCRab T cells, we next per-
major marker genes, GSEA analysis indicated a gene expression formed functional characterizations of effector cell properties.
profile similar to that of memory T cells rather than naive or Compared to iPSC-OP9-T cells or iPSC-SF-T cells, EZ-T cells
effector T cells (Figure 5D). None of the clusters showed sub- showed more robust upregulation of CD69 in response to
stantial expression of inhibitory receptors or regulatory T cell PMA/ionomycin treatment (Figure 6A). Consistently, EZ-T cells
markers (Figure S4B). expressed higher levels of CD107a than control iPSC-SF-T cells
We next sought to capture the compositional changes in im- upon PMA/ionomycin stimulation; the degranulation efficiency
mune cell clusters during T cell expansion. Compared with unac- was comparable to peripheral blood-derived T cells (Figure 6B).
tivated cells, T cell expansion resulted in enrichment of mature In light of the EZ-T cells’ superior activation/degranulation effi-
T cell populations and a concomitant reduction of immature ciency, we next explored whether EZ-T cells could be used to
DP T cells and innate-like cells (NK-like cells, gd T-like cells). generate CAR T cells with enhanced cytotoxic effector functions.
Similarly, naive-like T cells were predominantly present in unac- We transduced control iPSC-OP9-T cells, iPSC-SF-T cells, EZ-T
tivated samples and significantly reduced upon activation. cells, and donor-derived peripheral blood T cells with anti-CD19
Importantly, T cells displaying a memory-like phenotype were CARs containing a 4-1BB costimulatory domain, and co-
exclusively detected in activated cells after extended expansion, cultured with two different types of CD19+ lymphoma cells to
suggesting the occurrence of cell fate conversions from naive- compare their cytotoxicity profiles. CD19 CAR EZ-T cells caused
like cells into more differentiated subsets (Figures 5E and 5F). more efficient specific target cytolysis against both Jeko-1 and
To examine this, we performed gene regulatory network (GRN) OCI-Ly1 cells than CAR T cells derived from control iPSC-OP-
analysis using CellRouter, aiming to reconstruct the trajectories 9 and iPSC-SF-T cells and displayed a specific killing capacity
of cell-state transitions from naive to memory-like T cells (Lum- comparable to PBMC-T cells (Figures 6C and 6D). Cytotoxicity
mertz da Rocha et al., 2018). We identified a series of key tran- assays were also performed using presorted iPSC-SF TCRab
scriptional networks governing the changes in T cell composition T cells with or without EZH1 knockdown. Similarly, TCRab
and found multiple top-ranked transcriptional regulators, T cells with EZH1 knockdown elicited more efficient killing of
including BATF, LYAR, LITAF, IRF4, and RUNX3, which were target tumor cells (Figure S6A). Moreover, co-culture with tumor
previously known to drive the development of long-lived memory cells triggered EZ-T cells to secrete higher levels of cytokines
T cells (Figure 5G) (Seo et al., 2021; Chen et al., 2020; Mackay that are essential for T cell antitumor responses, including IL-2,
et al., 2013; Harberts et al., 2021; Wang et al., 2018b). Addition- interferon-g (IFN-g), and tumor necrosis factor a (TNFa)
ally, we performed single-cell regulatory network inference and (Figures 6E–6G). Both control SF-T and EZ-T cells produced
clustering (SCENIC) analysis to map regulons enriched in each lower levels of cytokines than PBMC-T cells. Because iPSC-
annotated cell type (Aibar et al., 2017) (Figure S4C). Consistent derived T cells are predominantly CD8+ cytotoxic cells, whereas
with the CellRouter analysis, SCENIC identified a similar set of the PBMC-T cells include a large proportion of CD4+ cells, this
regulons associated with the memory-like subset. In particular, observation is consistent with the higher cytokine production ca-
regulatory networks that have been linked to the generation pacity/profile of CD4 T cells (Pfizenmaier et al., 1984; Ngai et al.,
and homeostasis of memory T cells, such as BATF and IRF9 2007). Collectively, these data indicate that EZ-T cells exhibit
regulons, were exclusively enriched in the memory-like cluster enhanced cytotoxic and cytokine-producing effector functions
(Kurachi et al., 2014; Martinet et al., 2015; Seo et al., 2021) (Fig- against tumor cells in vitro.
ure S4D). Finally, we compared EZ-T cells with a range of he- To further evaluate the efficacy of EZ-T cell-derived CAR
matopoietic cells by mapping our scRNA-seq data on a publicly T cells, we established a xenograft mouse model by intrave-
available reference dataset that includes hematopoietic stem nously injecting luciferase-expressing diffuse large B cell lym-
cells (HSCs), lineage-restricted blood progenitors, and terminally phoma (DLBCL) cells (OCI-Ly1) into immunodeficient non-obese
differentiated lymphoid/erythroid/myeloid cells collected from diabetic-SCID IL2Rgammanull (NSG) mice. These animals were
healthy bone marrow and peripheral blood samples (Granja then infused with PBS, CD19 CAR iPSC-SF-T cells, CD19 CAR

ll
Article
A B
C D
E F G
Figure 6. EZ-T cells display enhanced effector functions

(A and B) (A) CD69 expression and (B) CD107a degranulation of iPSC-derived T cells and peripheral blood T cells after 6 h of PMA/ionomycin stimulation,
determined by flow cytometry analysis (n = 3, mean ± SEM).
(C and D) (C) CAR T cells generated from iPSC-derived T cells or peripheral blood T cells were co-cultured with JeKo1 and (D) OCI-Ly1 tumor cell lines at indicated
effector to target (E:T) ratios. Bar graph showing percentages of specific cytolysis of target tumor cells (n = 3, mean ± SEM).
(E–G) (E) Production of IL-2, (F) INFg, and (G) TNFa by CD19 CAR T cells generated from iPSC-SF-T, iPSC-EZ-T, and PBMC-T cells cultured in the absence (Un-
stim) and presence of OCI-Ly1 target tumor cells (n = 3, mean ± SEM, **** p % 0.0001).
EZ-T cells, or PBMC-derived CD19 CAR T cells, and subjected cause abnormal expansion of EZ-T cells, we administrated
to weekly bioluminescence imaging (BLI) to assess tumor both control and EZ-T cells into healthy mice and monitored
burden (Figure 7A). Although CAR T cells generated from the presence of CAR T cells over a long time period. Without
iPSC-SF-T cells suppressed tumor growth, they failed to eradi- simultaneous tumor engraftment, CAR T cells were barely
cate tumor cells in any animal after 7 weeks. Notably, CAR detectable in peripheral blood after 5 weeks, and these animals
EZ-T cells displayed significant improvement of efficacy and remained healthy and were free of human T cells after 20 weeks
were capable of eradicating tumor cells and caused complete (Figure S7B). In summary, when engineered with anti-CD19
remissions in most animals (Figure 7B). Consistent with CARs, EZ-T cells elicit superior antitumor effects relative to con-
improved tumor clearance, more CAR EZ-T cells were detected trol iPSC-SF-T cells lacking EZH1 knockdown, both in vitro and
in peripheral blood than control CAR iPSC-SF-T cells 3 weeks af- in vivo.
ter injection (Figure 7C), and most circulating EZ-T cells ex-
pressed abTCR and not gd TCR (Figure S6B), suggesting that DISCUSSION
the enhanced persistence of EZ-T cells is largely due to the
enrichment of TCRab T cells. As a result, though some mice The identification of signaling pathways that are essential for
showed persistent BLI signal intensity, the injection of CAR T cell development has led to the design of protocols that allow
EZ-T cells produced comparable survival rates to PBMC CAR the generation of T cells in vitro from human pluripotent stem
T cell treatment for the 7-week duration of the experiment cells (Holmes and Zúñiga-Pflu€cker, 2009; Timmermans et al.,
(Figures 7D and S7A). To exclude that EZH1 repression may 2009; Seet et al., 2017). However, past in vitro differentiation

ll
Article
C
D
(legend on next page)

ll
Article
approaches have been plagued by dependency on mouse stro- similar to T cells differentiated from CB CD34+ HSPCs. More-
mal cells as well as deficiencies in the terminal stages of T cell over, by avoiding the cumbersome co-culture with mouse feeder
maturation, thereby limiting their clinical application in adoptive cells, the stroma-free protocol minimizes batch-to-batch varia-
immunotherapy. The generation of definitive (adult-type) HSCs tion that can confound phenotypic characterization. Leveraging
with lymphoid potential and differentiation of functional T cells this stroma-free differentiation platform, we further demonstrate
from iPSCs has proven difficult, as in vitro differentiation of iPSCs that repression of EZH1 expression during lymphoid specifica-
tends to default into the production of embryonic cell types (Dou- tion facilitates T cell differentiation from human iPSCs and leads
latov et al., 2013; Sugimura et al., 2017). Past studies have to robust generation of developmentally mature iPSC-derived
demonstrated that iPSC-derived T cells resemble innate-like T cells. Importantly, while iPSC-SF-T cells exhibit some innate-
gd T cells and are not as robustly functional as primary ab like phenotypes that have been previously reported in OP9-
T cells, again suggesting that our ability to recapitulate the mech- stroma dependent iPSC differentiation systems (Themeli et al.,
anisms underlying commitment and progression of T cell devel- 2013), EZ-T cells display molecular signatures resembling pe-
opment is incomplete. Screening of critical epigenetic modifying ripheral blood TCRab T cells. These results are in agreement
enzymes during the generation of HSPCs discovered that EZH1 with the hypothesis that EZH1 knockdown produces definitive
plays a central role in regulating multipotency and lymphoid po- progenitors that preferentially support adult-like lymphopoiesis
tential in embryonic blood progenitors (Vo et al., 2018). As a rather than the generation of immature primitive embryonic
component of PRC2, EZH1 modulates chromatin accessibility lymphoid cells. Such a mechanism has also been supported
by mediating histone H3 lys27 trimethylation (H3K27me3) by recent findings that hPSC-derived CD34+ progenitors with a
(Shen et al., 2008). During embryonic hematopoiesis, EZH1 re- primary phenotype, defined by restricted hematopoietic poten-
presses the transcription of genes associated with definitive tial, produce fetal-like NK cells via in vitro differentiation (Dege
hematopoietic fates, and EZH1 deficiency in genetically engi- et al., 2020). In contrast, the multipotent CD34+ progenitors
neered mice promotes the precocious emergence of bona fide with full lymphoid potential are capable of giving rise to adult-
HSC and lymphoid progenitors (Vo et al., 2018). A recent study like NK cells that are functionally distinct from their fetal counter-
in the zebrafish model further showed that EZH1 suppresses parts (Dege et al., 2020).
HSPC formation by regulating HE commitment. Specifically, Another focus of our study was to test the impact of EZH1
EZH1 enhances Notch signaling to facilitate arterial gene expres- repression on the functional properties of iPSC-derived T cells
sion at the expense of HE specification and HSPC development. and to evaluate the potential of using EZ-T cells for adoptive
As a result, the knockdown of EZH1 unlocks definitive hemato- T cell immunotherapy. We engineered iPSC-derived T cells
poiesis and leads to the enhanced production of multipotent with anti-CD19 CARs and assessed their antitumor capacities.
HSPCs with lymphoid potential (Soto et al., 2021). Compared with CAR T cells generated from iPSC-SF-T cells,
In light of the repressive function of EZH1 in lymphopoiesis, we CD19-CAR EZ-T cells exhibit superior antitumor activity,
investigated the impact of EZH1 knockdown during in vitro T cell measured by tumor cell killing and cytokine production against
differentiation. Although 3D thymic-like culture systems have different types of CD19+ tumor cells in vitro and elicit more effi-
been established to mimic mouse or human T cell development cient tumor clearance in a xenograft mouse model. Notably,
and yield iPSC-derived T cells that display a mature phenotype, scRNA-seq analysis revealed that EZ-T cells, after activation,
these methods rely on engineered stromal cells or mouse- give rise to a subset of T cells that express relatively low levels
derived fetal thymic cells (Vizcardo et al., 2018; Montel-Hagen of cytotoxicity genes and high levels of memory T cell signature
et al., 2019; Wang et al., 2022). Therefore, we developed a genes. Such an observation is consistent with previous findings
stroma-free system that supports efficient T cell differentiation that in vitro T cell stimulation allows faster effector/memory
without using mouse-derived feeder cells. Such a strategy based commitment than physiological T cell differentiation/expansion
on immobilized Notch ligands has recently been employed to (Li and Kurlander, 2010; McLellan and Ali Hosseini Rad, 2019).
induce the iPSC-derived CD34+ HSPCs to differentiate into Trajectory analyses based on the activity of GRNs also identified
CD3+TCRab+CD8ab T cells that are immunophenotypically transcriptional regulators that drive the conversion of naive-like
similar to our iPSC-SF-T cells (Iriguchi et al., 2021; Trotman- EZ-T cells into memory-like cells. Among these networks are
Grant et al., 2021). However, whether such stroma-free differen- BATF, IRF4/9, and RUNX3, all of which are known to promote
tiation could support normal TCR rearrangement remained un- T cell longevity or prevent exhaustion (Martinet et al., 2015;
clear. Here, we show that the stroma-free system can faithfully Huber et al., 2017; Wang et al., 2018b; Istaces et al., 2019;
recapitulate T cell development by differentiating non-T cell- Seo et al., 2021). This is of particular interest as a growing
derived iPSCs (without pre-rearranged TCRs) into T cells with a body of evidence has shown that the enrichment of memory
high degree of TCR diversity. Compared with iPSC-OP9-T cells, T cells rather than terminally differentiated effector cells corre-
iPSC-SF-T cells exhibit a marginally more mature phenotype, lates with superior T cell persistence and improved clinical
Figure 7. CD19 CAR EZ-T cells mediate more robust in vivo tumor clearance
(A) Schematic illustration of in vivo CAR T cell functional studies using a DLBCL mouse model.
(B) Bioluminescent images of tumor xenografts over time and quantification of the tumor burden over time in each animal (n = 9 animals from 3 experiments),
represented by mean total flux (photons/s).
(C) Numbers of CAR T cells per 100 mL peripheral blood from each animal, determined by flow cytometry 3 weeks after CAR T cell injection (n = 8, *** p % 0.001).
(D) Kaplan-Meier curve showing percentage survival of untreated animals (black) and animal groups treated with CD19 CAR T cells generated from control iPSC-
SF-T (red), iPSC-EZ-T (blue), or PBMC-T cells (yellow) (n = 9, ** p % 0.01 by log-rank test).

ll
Article
outcomes in adoptive T cell therapies (Gattinoni et al., 2005; Kle- STAR+METHODS

banoff et al., 2005; Stark et al., 2009; Fraietta et al., 2018a,
2018b). Therefore, the presence of a memory-like T cell subset Detailed methods are provided in the online version of this paper
may contribute to the enhanced persistence and antitumor activ- and include the following:
ity of EZ-T cells in the DLBCL model.
New cell engineering strategies have profoundly changed the d KEY RESOURCES TABLE
paradigm of CAR T cell therapy. Multiple studies have generated d RESOURCE AVAILABILITY
CAR T cells with disrupted endogenous TCR to avoid the risk of B Lead contact
graft-versus-host disease (GVHD) inherent in allogeneic T cell B Materials availability

B Data and code availability
therapy (MacLeod et al., 2017; Eyquem et al., 2017; Georgiadis
et al., 2018). Recent success in producing hypoimmunogenic d EXPERIMENTAL MODEL AND SUBJECT DETAILS
iPSCs that can evade immune rejection has further enhanced B Cell line
B In vivo tumor xenograft model
the prospects for off-the-shelf, universal iPSC-derived CAR
T cells (Han et al., 2019; Deuse et al., 2019; Wang et al., 2021). d METHOD DETAILS
+
Given the fact that our stroma-free system produces iPSC- B Culture of iPSCs and generation of CD34 HE
B Differentiation of iPSCs into T cells
derived T cells expressing a highly diverse TCR repertoire, ge-
B CRISPR interference in iPSs
netic ablation of the endogenous TCR will be required before
allogeneic transplantation. Compared with primary T cells, B Immunoblot
iPSCs are more amenable to genetic manipulations and could B Quantitative Real-Time PCR analysis
B Flow cytometry
facilitate the engineering of ‘‘armored’’ CAR T cells that can
secrete specific cytokines or checkpoint inhibitor antibodies to B Bulk RNA-seq and data analysis
overcome the suppressive tumor microenvironment (Pegram B TCR repertoire analysis

B Single-cell RNA-Sequencing and data analysis
et al., 2015; Adachi et al., 2018; Rafiq et al., 2018). In addition
B T cell activation and degranulation
to T cell-based immunotherapies, NK cells have also shown
great promise in the treatment of both blood and solid tumors B CAR T cell functional assays
and are currently being tested in multiple clinical trials (Basar d QUANTIFICATION AND STATISTICAL ANALYSIS
et al., 2020; Liu et al., 2021). Compared with T cells, NK cell cyto-
SUPPLEMENTAL INFORMATION
toxicity is not constrained by MHC recognition and could target
tumor cells that are resistant to T cell killing. Moreover, NK cells Supplemental information can be found online at https://doi.org/10.1016/j.
are less prone to GVHD and CAR-associated toxicity (Ruggeri stem.2022.06.014.
et al., 2002; Liu et al., 2020). Multiple studies have successfully
generated iPSC-derived NK cells that elicit antitumor activities ACKNOWLEDGMENTS
(Knorr et al., 2013; Zeng et al., 2017; Li et al., 2018; Woan
We thank the Boston Children’s Hospital Flow Cytometry Core and the hESC
et al., 2021). How EZH1-mediated regulation of lymphoid
Core Facility. We also thank John Atwater, Merve Akyol, and Tamer Onder for
commitment might affect NK cell differentiation from iPSCs re- assistance with experiments and advice. This study was supported by grants
mains an open question. from NIH NHLBI U01HL134812 (G.Q.D.), NIH NIDDK 2U01DK104218 (G.Q.D.),
NIH NHLBI 5U24HL134763 subaward 1701192 (R.J.), the Emerson Collective
Limitations of the study Cancer Research Fund (G.Q.D.), and philanthropic support from the Manton
Although our data suggest that EZ-T cells show comparable Foundation.
efficacy to donor-derived CAR T cells, there are limitations
AUTHOR CONTRIBUTIONS
to be addressed in future studies. First, iPSC-derived EZ-T
cells produced by existing methods predominantly consist R.J. and G.Q.D. conceived the project and designed the experiments. R.J.
of CD8 cytotoxic T cells. Therefore, new strategies are needed performed iPSC and iPSC-T cell studies. R.J. and I.S. performed CAR T cell
functional assays. M.N., E.L.d.R., and M.F. analyzed scRNA-seq data. A.H.
to enable the efficient production of mature CD4 SP T cells, as
and M.N. analyzed bulk RNA-seq data. T.B. and M.N. generated CRISPRi
a more balanced ratio of cytotoxic versus helper T cells has
iPSC lines. R.J. and M.S. performed studies in the xenograft mouse model.
been associated with improved therapeutic responses (Som- C.K. and D.J. participated in experimental optimization and data interpreta-
mermeyer et al., 2016; Wang et al., 2018a). Second, our cur- tion. T.M.S., T.E.N., M.V.M., and G.Q.D. participated in project planning, su-
rent methods for repressing EZH1 activity during T cell differ- pervised research, and interpreted data. R.J. and G.Q.D. wrote the manuscript
entiation employ viral vectors that integrate into host DNA. with input from all co-authors.
Though EZH1/2 dual inhibitors have been shown to promote
DECLARATION OF INTERESTS
NK cell development (Damele et al., 2021), small molecule
treatment fails to phenocopy shRNA-mediated EZH1 knock- R.J., G.Q.D., and Boston Children’s Hospital hold intellectual property and
down during iPSC-T cell differentiation. This is consistent receive consulting fees and/or hold equity interest relevant to the generation
with previous findings that EZH1/EZH2 double knockdown of iPSC-derived T cells. T.M.S. receives sponsored research support from
Elevate Bio. G.Q.D. is a member of Cell Stem Cell’s advisory board.
had no effects on T cell potential (Vo et al., 2018). Given the
lack of EZH1-specific inhibitors that do not repress EZH2
Received: February 2, 2022
function, the development of non-integrating gene knockdown Revised: May 31, 2022
strategies will be needed to facilitate translation and clinical Accepted: June 29, 2022
applications of EZ-T cells. Published: August 4, 2022

ll
Article
REFERENCES Fesnak, A.D., June, C.H., and Levine, B.L. (2016b). Engineered T cells: the
promise and challenges of cancer immunotherapy. Nat. Rev. Cancer 16,
Adachi, K., Kano, Y., Nagai, T., Okuyama, N., Sakoda, Y., and Tamada, K. 566–581.
(2018). IL-7 and CCL19 expression in CAR-T cells improves immune cell infil- Fraietta, J.A., Lacey, S.F., Orlando, E.J., Pruteanu-Malinici, I., Gohil, M.,
tration and CAR-T cell survival in the tumor. Nat. Biotechnol. 36, 346–351. Lundh, S., Boesteanu, A.C., Wang, Y., O’Connor, R.S., Hwang, W.T., et al.
Aibar, S., González-Blas, C.B., Moerman, T., Huynh-Thu, V.A., Imrichova, H., (2018a). Determinants of response and resistance to CD19 chimeric antigen
Hulselmans, G., Rambow, F., Marine, J.C., Geurts, P., Aerts, J., et al. (2017). receptor (CAR) T cell therapy of chronic lymphocytic leukemia. Nat. Med.
SCENIC: single-cell regulatory network inference and clustering. Nat. 24, 563–571.
Methods 14, 1083–1086. Fraietta, J.A., Nobles, C.L., Sammons, M.A., Lundh, S., Carty, S.A., Reich, T.J.,
Baldwin, T.A., Sandau, M.M., Jameson, S.C., and Hogquist, K.A. (2005). The Cogdill, A.P., Morrissette, J.J.D., DeNizio, J.E., Reddy, S., et al. (2018b).
timing of TCR alpha expression critically influences T cell development and se- Disruption of TET2 promotes the therapeutic efficacy of CD19-targeted
lection. J. Exp. Med. 202, 111–121. T cells. Nature 558, 307–312.
Basar, R., Daher, M., and Rezvani, K. (2020). Next-generation cell therapies: Gattinoni, L., Klebanoff, C.A., Palmer, D.C., Wrzesinski, C., Kerstann, K., Yu,
the emerging role of CAR-NK cells. Hematology Am. Soc. Hematol. Educ. Z., Finkelstein, S.E., Theoret, M.R., Rosenberg, S.A., and Restifo, N.P.
Program 2020, 570–578. (2005). Acquisition of full effector function in vitro paradoxically impairs the
Bosselut, R., Kubo, S., Guinter, T., Kopacz, J.L., Altman, J.D., Feigenbaum, L., in vivo antitumor efficacy of adoptively transferred CD8+ T cells. J. Clin.
and Singer, A. (2000). Role of CD8beta domains in CD8 coreceptor function: Invest. 115, 1616–1626.
importance for MHC I binding, signaling, and positive selection of CD8+ Georgiadis, C., Preece, R., Nickolay, L., Etuk, A., Petrova, A., Ladon, D., Danyi,
T cells in the thymus. Immunity 12, 409–418. A., Humphryes-Kirilov, N., Ajetunmobi, A., Kim, D., et al. (2018). Long terminal
Cahan, P., Li, H., Morris, S.A., Lummertz da Rocha, E., Daley, G.Q., and repeat CRISPR-CAR-coupled ‘‘universal’’ T cells mediate potent anti-leukemic
Collins, J.J. (2014). CellNet: network biology applied to stem cell engineering. effects. Mol. Ther. 26, 1215–1227.
Cell 158, 903–915. Good, C.R., Aznar, M.A., Kuramitsu, S., Samareh, P., Agarwal, S., Donahue,
Chang, C.W., Lai, Y.S., Lamb, L.S., and Townes, T.M. (2014). Broad T-cell re- G., Ishiyama, K., Wellhausen, N., Rennels, A.K., Ma, Y., et al. (2021). An NK-
ceptor repertoire in T-lymphocytes derived from human induced pluripotent like CAR T cell transition in CAR T cell dysfunction. Cell 184, 6081–6100.e26.
stem cells. PLoS One 9, e97335. Granja, J.M., Klemm, S., McGinnis, L.M., Kathiria, A.S., Mezger, A., Corces,
Chen, Q.Y., Li, Y.N., Wang, X.Y., Zhang, X., Hu, Y., Li, L., Suo, D.Q., Ni, K., Li, M.R., Parks, B., Gars, E., Liedtke, M., Zheng, G.X.Y., et al. (2019). Single-
Z., Zhan, J.R., et al. (2020). Tumor fibroblast-derived FGF2 regulates expres- cell multiomic analysis identifies regulatory programs in mixed-phenotype
sion of SPRY1 in esophageal tumor-infiltrating T cells and plays a role in T-cell acute leukemia. Nat. Biotechnol. 37, 1458–1465.
exhaustion. Cancer Res. 80, 5583–5596. Han, X., Wang, M., Duan, S., Franco, P.J., Kenty, J.H., Hedrick, P., Xia, Y.,
Cheroutre, H., and Lambolez, F. (2008). Doubting the TCR coreceptor function Allen, A., Ferreira, L.M.R., Strominger, J.L., et al. (2019). Generation of hypoim-
of CD8alphaalpha. Immunity 28, 149–159. munogenic human pluripotent stem cells. Proc. Natl. Acad. Sci. USA 116,
10441–10446.
Damele, L., Amaro, A., Serio, A., Luchetti, S., Pfeffer, U., Mingari, M.C., and
Vitale, C. (2021). EZH1/2 inhibitors favor ILC3 development from human €cker,
Harberts, A., Schmidt, C., Schmid, J., Reimers, D., Koch-Nolte, F., Mittru
HSPC-CD34+ cells. Cancers (Basel) 13, 319. H.W., and Raczkowski, F. (2021). Interferon regulatory factor 4 controls
effector functions of CD8+ memory T cells. Proc. Natl. Acad. Sci. USA 118.
Dang, Y., Jia, G., Choi, J., Ma, H., Anaya, E., Ye, C., Shankar, P., and Wu, H.
e2014553118.
(2015). Optimizing sgRNA structure to improve CRISPR-Cas9 knockout effi-
ciency. Genome Biol. 16, 280. Holler, P.D., and Kranz, D.M. (2003). Quantitative analysis of the contribution of
Dege, C., Fegan, K.H., Creamer, J.P., Berrien-Elliott, M.M., Luff, S.A., Kim, D., TCR/pepMHC affinity and CD8 to T cell activation. Immunity 18, 255–264.
Wagner, J.A., Kingsley, P.D., McGrath, K.E., Fehniger, T.A., et al. (2020). €cker, J.C. (2009). The OP9-DL1 system: genera-
Holmes, R., and Zúñiga-Pflu
Potently cytotoxic natural killer cells initially emerge from erythro-myeloid pro- tion of T-lymphocytes from embryonic or hematopoietic stem cells in vitro.
genitors during mammalian development. Dev. Cell 53, 229–239.e7. Cold Spring Harb. Protoc. 2009. pdb.prot5156.
Deuse, T., Hu, X., Gravina, A., Wang, D., Tediashvili, G., De, C., Thayer, W.O., Huber, M., Suprunenko, T., Ashhurst, T., Marbach, F., Raifer, H., Wolff, S.,
Wahl, A., Garcia, J.V., Reichenspurner, H., et al. (2019). Hypoimmunogenic de- Strecker, T., Viengkhou, B., Jung, S.R., Obermann, H.L., et al. (2017). IRF9 pre-
rivatives of induced pluripotent stem cells evade immune rejection in fully vents CD8+ T cell exhaustion in an extrinsic manner during acute lymphocytic
immunocompetent allogeneic recipients. Nat. Biotechnol. 37, 252–258. choriomeningitis virus infection. J. Virol. 91. e01219-17.
Ditadi, A., Sturgeon, C.M., Tober, J., Awong, G., Kennedy, M., Yzaguirre, A.D., Huijskens, M.J., Walczak, M., Koller, N., Briedé, J.J., Senden-Gijsbers, B.L.,
Azzola, L., Ng, E.S., Stanley, E.G., French, D.L., et al. (2015). Human definitive Schnijderberg, M.C., Bos, G.M., and Germeraad, W.T. (2014). Technical
haemogenic endothelium and arterial vascular endothelium represent distinct advance: ascorbic acid induces development of double-positive T cells from
lineages. Nat. Cell Biol. 17, 580–591. human hematopoietic stem cells in the absence of stromal cells. J. Leukoc.
Doulatov, S., Vo, L.T., Chou, S.S., Kim, P.G., Arora, N., Li, H., Hadland, B.K., Biol. 96, 1165–1175.
Bernstein, I.D., Collins, J.J., Zon, L.I., and Daley, G.Q. (2013). Induction of Huster, K.M., Busch, V., Schiemann, M., Linkemann, K., Kerksiek, K.M.,
multipotential hematopoietic progenitors from human pluripotent stem cells Wagner, H., and Busch, D.H. (2004). Selective expression of IL-7 receptor
via respecification of lineage-restricted precursors. Cell Stem Cell 13, on memory T cells identifies early CD40L-dependent generation of distinct
459–470. CD8+ memory T cell subsets. Proc. Natl. Acad. Sci. USA 101, 5610–5615.
Egawa, T., Kreslavsky, T., Littman, D.R., and von Boehmer, H. (2008). Lineage Intlekofer, A.M., Takemoto, N., Wherry, E.J., Longworth, S.A., Northrup, J.T.,
diversion of T cell receptor transgenic thymocytes revealed by lineage fate Palanivel, V.R., Mullen, A.C., Gasink, C.R., Kaech, S.M., Miller, J.D., et al.
mapping. PLoS One 3, e1512. (2005). Effector and memory CD8+ T cell fate coupled by T-bet and eomeso-
Eyquem, J., Mansilla-Soto, J., Giavridis, T., van der Stegen, S.J., Hamieh, M., dermin. Nat. Immunol. 6, 1236–1244.
Cunanan, K.M., Odak, A., Gönen, M., and Sadelain, M. (2017). Targeting a Iriguchi, S., Yasui, Y., Kawai, Y., Arima, S., Kunitomo, M., Sato, T., Ueda, T.,
CAR to the TRAC locus with CRISPR/Cas9 enhances tumour rejection. Minagawa, A., Mishima, Y., Yanagawa, N., et al. (2021). A clinically applicable
Nature 543, 113–117. and scalable method to regenerate T-cells from iPSCs for off-the-shelf T-cell
Fesnak, A., Lin, C.Y., Siegel, D.L., and Maus, M.V. (2016a). CAR-T cell thera- immunotherapy. Nat. Commun. 12, 430.
pies from the transfusion medicine perspective. Transfus. Med. Rev. 30, Istaces, N., Splittgerber, M., Lima Silva, V., Nguyen, M., Thomas, S., Le, A.,
139–145. Achouri, Y., Calonne, E., Defrance, M., Fuks, F., et al. (2019). EOMES interacts

ll
Article
with RUNX3 and BRG1 to promote innate memory cell formation through Martinet, V., Tonon, S., Torres, D., Azouz, A., Nguyen, M., Kohler, A., Flamand,
epigenetic reprogramming. Nat. Commun. 10, 3306. V., Mao, C.A., Klein, W.H., Leo, O., and Goriely, S. (2015). Type I interferons
June, C.H., O’Connor, R.S., Kawalekar, O.U., Ghassemi, S., and Milone, M.C. regulate eomesodermin expression and the development of unconventional
(2018). CAR T cell immunotherapy for human cancer. Science 359, memory CD8(+) T cells. Nat. Commun. 6, 7089.
1361–1365. Mathewson, N.D., Ashenberg, O., Tirosh, I., Gritsch, S., Perez, E.M., Marx, S.,
Kang, J.B., Nathan, A., Weinand, K., Zhang, F., Millard, N., Rumker, L., Moody, Jerby-Arnon, L., Chanoch-Myers, R., Hara, T., Richman, A.R., et al. (2021).
D.B., Korsunsky, I., and Raychaudhuri, S. (2021). Efficient and precise single- Inhibitory CD161 receptor identified in glioma-infiltrating T cells by single-
cell reference atlas mapping with Symphony. Nat. Commun. 12, 5890. cell analysis. Cell 184, 1281–1298.e26.
Klebanoff, C.A., Gattinoni, L., Torabi-Parizi, P., Kerstann, K., Cardones, A.R., McLellan, A.D., and Ali Hosseini Rad, S.M. (2019). Chimeric antigen receptor
Finkelstein, S.E., Palmer, D.C., Antony, P.A., Hwang, S.T., Rosenberg, S.A., T cell persistence and memory cell formation. Immunol. Cell Biol. 97, 664–674.
et al. (2005). Central memory self/tumor-reactive CD8+ T cells confer superior Montel-Hagen, A., and Crooks, G.M. (2019). From pluripotent stem cells to
antitumor immunity compared with effector memory T cells. Proc. Natl. Acad. T cells. Exp. Hematol. 71, 24–31.
Sci. USA 102, 9571–9576.
Montel-Hagen, A., Seet, C.S., Li, S., Chick, B., Zhu, Y., Chang, P., Tsai, S.,
Knorr, D.A., Ni, Z., Hermanson, D., Hexum, M.K., Bendzick, L., Cooper, L.J., Sun, V., Lopez, S., Chen, H.C., et al. (2019). Organoid-induced differentiation
Lee, D.A., and Kaufman, D.S. (2013). Clinical-scale derivation of natural killer of conventional T cells from human pluripotent stem cells. Cell Stem Cell 24,
cells from human pluripotent stem cells for cancer therapy. Stem Cells 376–389.e8.
Transl. Med. 2, 274–283.
Ngai, P., McCormick, S., Small, C., Zhang, X., Zganiacz, A., Aoki, N., and Xing,
Kumar, B.V., Connors, T.J., and Farber, D.L. (2018). Human T cell develop- Z. (2007). Gamma interferon responses of CD4 and CD8 T-cell subsets are
ment, localization, and function throughout life. Immunity 48, 202–213. quantitatively different and independent of each other during pulmonary
Kurachi, M., Barnitz, R.A., Yosef, N., Odorizzi, P.M., DiIorio, M.A., Lemieux, Mycobacterium bovis BCG infection. Infect. Immun. 75, 2244–2252.
M.E., Yates, K., Godec, J., Klatt, M.G., Regev, A., et al. (2014). The transcrip- Nishimura, T., Kaneko, S., Kawana-Tachikawa, A., Tajima, Y., Goto, H., Zhu,
tion factor BATF operates as an essential differentiation checkpoint in early D., Nakayama-Hosoya, K., Iriguchi, S., Uemura, Y., Shimizu, T., et al. (2013).
effector CD8+ T cells. Nat. Immunol. 15, 373–383. Generation of rejuvenated antigen-specific T cells by reprogramming to plurip-
€cker, J.C. (2005). Induction of
La Motte-Mohs, R.N., Herer, E., and Zúñiga-Pflu otency and redifferentiation. Cell Stem Cell 12, 114–126.
T-cell development from human cord blood hematopoietic stem cells by delta- Pegram, H.J., Purdon, T.J., van Leeuwen, D.G., Curran, K.J., Giralt, S.A.,
like 1 in vitro. Blood 105, 1431–1439. Barker, J.N., and Brentjens, R.J. (2015). IL-12-secreting CD19-targeted cord
Li, Y., Hermanson, D.L., Moriarity, B.S., and Kaufman, D.S. (2018). Human blood-derived T cells for the immunotherapy of B-cell acute lymphoblastic leu-
iPSC-derived natural killer cells engineered with chimeric antigen receptors kemia. Leukemia 29, 415–422.
enhance anti-tumor activity. Cell Stem Cell 23, 181–192.e5. €ubener, W., Krönke, M., Röllinghoff, M., and
Pfizenmaier, K., Scheurich, P., Da
Li, Y., and Kurlander, R.J. (2010). Comparison of anti-CD3 and anti-CD28- Wagner, H. (1984). Quantitative representation of all T cells committed to
coated beads with soluble anti-CD3 for expanding human T cells: differing develop into cytotoxic effector cells and/or interleukin 2 activity-producing
impact on CD8 T cell phenotype and responsiveness to restimulation. helper cells within murine T lymphocyte subsets. Eur. J. Immunol. 14, 33–39.
J. Transl. Med. 8, 104. Radley, A.H., Schwab, R.M., Tan, Y., Kim, J., Lo, E.K.W., and Cahan, P. (2017).
Liu, E., Marin, D., Banerjee, P., Macapinlac, H.A., Thompson, P., Basar, R., Assessment of engineered cells using CellNet and RNA-seq. Nat. Protoc. 12,
Nassif Kerbauy, L., Overman, B., Thall, P., Kaplan, M., et al. (2020). Use of 1089–1102.
CAR-transduced natural killer cells in CD19-positive lymphoid tumors. Rafiq, S., Yeku, O.O., Jackson, H.J., Purdon, T.J., van Leeuwen, D.G., Drakes,
N. Engl. J. Med. 382, 545–553. D.J., Song, M., Miele, M.M., Li, Z., Wang, P., et al. (2018). Targeted delivery of a
Liu, S., Galat, V., Galat, Y., Lee, Y.K.A., Wainwright, D., and Wu, J. (2021). NK PD-1-blocking scFv by CAR-T cells enhances anti-tumor efficacy in vivo. Nat.
cell-based cancer immunotherapy: from basic biology to clinical development. Biotechnol. 36, 847–856.
J. Hematol. Oncol. 14, 7. Ruggeri, L., Capanni, M., Urbani, E., Perruccio, K., Shlomchik, W.D., Tosti, A.,
Lummertz da Rocha, E., Kubaczka, C., Sugden, W.W., Najia, M.A., Jing, R., Posati, S., Rogaia, D., Frassoni, F., Aversa, F., et al. (2002). Effectiveness of
Markel, A., LeBlanc, Z.C., Dos Santos Peixoto, R., Falchetti, M., Collins, J.J., donor natural killer cell alloreactivity in mismatched hematopoietic transplants.
et al. (2022). CellComm infers cellular crosstalk that drives haematopoietic Science 295, 2097–2100.
stem and progenitor cell development. Nat. Cell Biol. 24, 579–589. Sanders, M.E., Makgoba, M.W., Sharrow, S.O., Stephany, D., Springer, T.A.,
Lummertz da Rocha, E., Rowe, R.G., Lundin, V., Malleshaiah, M., Jha, D.K., Young, H.A., and Shaw, S. (1988). Human memory T lymphocytes express
Rambo, C.R., Li, H., North, T.E., Collins, J.J., and Daley, G.Q. (2018). increased levels of three cell adhesion molecules (LFA-3, CD2, and LFA-1)
Reconstruction of complex single-cell trajectories using CellRouter. Nat. and three other molecules (UCHL1, CDw29, and Pgp-1) and have enhanced
Commun. 9, 892. IFN-gamma production. J. Immunol. 140, 1401–1407.
Mackay, L.K., Rahimpour, A., Ma, J.Z., Collins, N., Stock, A.T., Hafon, M.L., Sanson, K.R., Hanna, R.E., Hegde, M., Donovan, K.F., Strand, C., Sullender,
Vega-Ramos, J., Lauzurica, P., Mueller, S.N., Stefanovic, T., et al. (2013). M.E., Vaimberg, E.W., Goodale, A., Root, D.E., Piccioni, F., and Doench,
The developmental pathway for CD103(+)CD8+ tissue-resident memory J.G. (2018). Optimized libraries for CRISPR-Cas9 genetic screens with multiple
T cells of skin. Nat. Immunol. 14, 1294–1301. modalities. Nat. Commun. 9, 5416.
MacLeod, D.T., Antony, J., Martin, A.J., Moser, R.J., Hekele, A., Wetzel, K.J., Scarfò, I., Ormhøj, M., Frigault, M.J., Castano, A.P., Lorrey, S., Bouffard, A.A.,
Brown, A.E., Triggiano, M.A., Hux, J.A., Pham, C.D., et al. (2017). Integration of van Scoyk, A., Rodig, S.J., Shay, A.J., Aster, J.C., et al. (2018). Anti-CD37
a CD19 CAR into the TCR alpha chain locus streamlines production of alloge- chimeric antigen receptor T cells are active against B- and T-cell lymphomas.
neic gene-edited CAR T cells. Mol. Ther. 25, 949–961. Blood 132, 1495–1506.
Maeda, T., Nagano, S., Ichise, H., Kataoka, K., Yamada, D., Ogawa, S., Seet, C.S., He, C., Bethune, M.T., Li, S., Chick, B., Gschweng, E.H., Zhu, Y.,
Koseki, H., Kitawaki, T., Kadowaki, N., Takaori-Kondo, A., et al. (2016). Kim, K., Kohn, D.B., Baltimore, D., et al. (2017). Generation of mature T cells
Regeneration of CD8ab T cells from T-cell-derived iPSC imparts potent tumor from human hematopoietic stem and progenitor cells in artificial thymic orga-
antigen-specific cytotoxicity. Cancer Res. 76, 6839–6850. noids. Nat. Methods 14, 521–530.
Marçais, A., Coupet, C.A., Walzer, T., Tomkowiak, M., Ghittoni, R., and Marvel, Seo, H., González-Avalos, E., Zhang, W., Ramchandani, P., Yang, C., Lio, C.J.,
J. (2006). Cell-autonomous CCL5 transcription by memory CD8 T cells is regu- Rao, A., and Hogan, P.G. (2021). BATF and IRF4 cooperate to counter exhaus-
lated by IL-4. J. Immunol. 177, 4451–4457. tion in tumor-infiltrating CAR T cells. Nat. Immunol. 22, 983–995.

ll
Article
Shen, X., Liu, Y., Hsu, Y.J., Fujiwara, Y., Kim, J., Mao, X., Yuan, G.C., and Vizcardo, R., Klemen, N.D., Islam, S.M.R., Gurusamy, D., Tamaoki, N.,
Orkin, S.H. (2008). EZH1 mediates methylation on histone H3 lysine 27 and Yamada, D., Koseki, H., Kidder, B.L., Yu, Z., Jia, L., et al. (2018). Generation
complements EZH2 in maintaining stem cell identity and executing pluripo- of tumor antigen-specific iPSC-derived thymic emigrants using a 3D thymic
tency. Mol. Cell 32, 491–502. culture system. Cell Rep. 22, 3175–3190.
Shukla, S., Langley, M.A., Singh, J., Edgar, J.M., Mohtashami, M., Zúñiga- Vizcardo, R., Masuda, K., Yamada, D., Ikawa, T., Shimizu, K., Fujii, S., Koseki,
€cker, J.C., and Zandstra, P.W. (2017). Progenitor T-cell differentiation
Pflu H., and Kawamoto, H. (2013). Regeneration of human tumor antigen-specific
from hematopoietic stem cells using Delta-like-4 and VCAM-1. Nat. T cells from iPSCs derived from mature CD8(+) T cells. Cell Stem Cell
Methods 14, 531–538. 12, 31–36.
Sommermeyer, D., Hudecek, M., Kosasih, P.L., Gogishvili, T., Maloney, D.G., Vo, L.T., Kinney, M.A., Liu, X., Zhang, Y., Barragan, J., Sousa, P.M., Jha, D.K.,
Turtle, C.J., and Riddell, S.R. (2016). Chimeric antigen receptor-modified Han, A., Cesana, M., Shao, Z., et al. (2018). Regulation of embryonic haema-
T cells derived from defined CD8+ and CD4+ subsets confer superior anti- topoietic multipotency by EZH1. Nature 553, 506–510.
tumor reactivity in vivo. Leukemia 30, 492–500. Wang, B., Iriguchi, S., Waseda, M., Ueda, N., Ueda, T., Xu, H., Minagawa, A.,
Soto, R.A., Najia, M.A.T., Hachimi, M., Frame, J.M., Yette, G.A., Lummertz da Ishikawa, A., Yano, H., Ishi, T., et al. (2021). Generation of hypoimmunogenic
Rocha, E., Stankunas, K., Daley, G.Q., North, T.E., Stankunas, K., Daley, G.Q., T cells from genetically engineered allogeneic human induced pluripotent
and North, T.E. (2021). Sequential regulation of hemogenic fate and hemato- stem cells. Nat. Biomed. Eng. 5, 429–440.
poietic stem and progenitor cell formation from arterial endothelium by Wang, D., Aguilar, B., Starr, R., Alizadeh, D., Brito, A., Sarkissian, A., Ostberg,
Ezh1/2. Stem Cell Reports 16, 1718–1734. J.R., Forman, S.J., and Brown, C.E. (2018a). Glioblastoma-targeted CD4+
Stark, F.C., Sad, S., and Krishnan, L. (2009). Intracellular bacterial vectors that CAR T cells mediate superior antitumor activity. JCI Insight 3, e99048.
induce CD8(+) T cells with similar cytolytic abilities but disparate memory phe- Wang, D., Diao, H., Getzler, A.J., Rogal, W., Frederick, M.A., Milner, J., Yu, B.,
notypes provide contrasting tumor protection. Cancer Res. 69, 4327–4334. Crotty, S., Goldrath, A.W., and Pipkin, M.E. (2018b). The transcription factor
Sugimura, R., Jha, D.K., Han, A., Soria-Valles, C., da Rocha, E.L., Lu, Y.F., Runx3 establishes chromatin accessibility of cis-regulatory landscapes that
Goettel, J.A., Serrao, E., Rowe, R.G., Malleshaiah, M., et al. (2017). drive memory cytotoxic T lymphocyte formation. Immunity 48, 659–674.e6.
Haematopoietic stem and progenitor cells from human pluripotent stem cells. Wang, Z., McWilliams-Koeppen, H.P., Reza, H., Ostberg, J.R., Chen, W.,
Nature 545, 432–438. Wang, X., Huynh, C., Vyas, V., Chang, W.C., Starr, R., et al. (2022). 3D-orga-
Terrence, K., Pavlovich, C.P., Matechak, E.O., and Fowlkes, B.J. (2000). noid culture supports differentiation of human CAR+ iPSCs into highly func-
Premature expression of T cell receptor (TCR)ab suppresses TCRgd gene re- tional CAR T cells. Cell Stem Cell 29, 515–527.e8.
arrangement but permits development of gammadelta lineage T cells. J. Exp. Willinger, T., Freeman, T., Herbert, M., Hasegawa, H., McMichael, A.J., and
Med. 192, 537–548. Callan, M.F. (2006). Human naive CD8 T cells down-regulate expression of
Themeli, M., Kloss, C.C., Ciriello, G., Fedorov, V.D., Perna, F., Gonen, M., and the WNT pathway transcription factors lymphoid enhancer binding factor 1
Sadelain, M. (2013). Generation of tumor-targeted human T lymphocytes from and transcription factor 7 (T cell factor-1) following antigen encounter in vitro
induced pluripotent stem cells for cancer therapy. Nat. Biotechnol. 31, and in vivo. J. Immunol. 176, 1439–1446.
928–933. Woan, K.V., Kim, H., Bjordahl, R., Davis, Z.B., Gaidarova, S., Goulding, J.,
Timmermans, F., Velghe, I., Vanwalleghem, L., De Smedt, M., Van Hancock, B., Mahmood, S., Abujarour, R., Wang, H., et al. (2021).
Coppernolle, S., Taghon, T., Moore, H.D., Leclercq, G., Langerak, A.W., Harnessing features of adaptive NK cells to generate iPSC-derived NK cells
Kerre, T., et al. (2009). Generation of T cells from human embryonic stem for enhanced immunotherapy. Cell Stem Cell 28, 2062–2075.e5.
cell-derived hematopoietic zones. J. Immunol. 182, 6879–6888. Zeng, J., Tang, S.Y., Toh, L.L., and Wang, S. (2017). Generation of ‘‘off-the-
Trotman-Grant, A.C., Mohtashami, M., De Sousa Casal, J., Martinez, E.C., shelf’’ natural killer cells from peripheral blood cell-derived induced pluripotent
Lee, D., Teichman, S., Brauer, P.M., Han, J., Anderson, M.K., and Zúñiga- stem cells. Stem Cell Rep. 9, 1796–1812.
€cker, J.C. (2021). DL4-mbeads induce T cell lineage differentiation from
Pflu Zhao, Y., Parkhurst, M.R., Zheng, Z., Cohen, C.J., Riley, J.P., Gattinoni, L.,
stem cells in a stromal cell-free system. Nat. Commun. 12, 5023. Restifo, N.P., Rosenberg, S.A., and Morgan, R.A. (2007). Extrathymic genera-
Vivier, E., and Anfossi, N. (2004). Inhibitory NK-cell receptors on T cells: wit- tion of tumor-specific T cells from genetically engineered human hematopoiet-
ness of the past, actors of the future. Nat. Rev. Immunol. 4, 190–198. ic stem cells via Notch signaling. Cancer Res. 67, 2425–2429.

ll
Article
STAR+METHODS
KEY RESOURCES TABLE
REAGENT or RESOURCE SOURCE IDENTIFIER

Antibodies
Anti-human CD45RA BV510 Thermo Fisher Scientific Cat#BDB563031; RRID:AB_2722499
Anti-human CD45RO PE Thermo Fisher Scientific Cat#BDB555493; RRID:AB_395884
Anti-human CCR7 APC BD Bioscience Cat#566762; RRID:AB_2869854
Anti-human CD3 PE/cy7 Biolegend Cat#344816; RRID:AB_10640737
Anti-human CD8 BV421 Thermo Fisher Scientific Cat#BDB562428; RRID:AB_11154035
Anti-human TCRab APC Biolegend Cat#306717; RRID:AB_10612747
Anti-human TCRgd PE Biolegend Cat#331209; RRID:AB_1089219
Anti-human CD7 PE Thermo Fisher Scientific Cat#BDB555361; RRID:AB_395764
Anti-human CD5 BV510 Thermo Fisher Scientific Cat#BDB563381; RRID:AB_2744435
Anti-human CD4 PE/cy5 Biolegend Cat#555348; RRID:AB_395753
Anti-human CD8beta APC Miltenyi Biotec Cat#130-110-569; RRID:AB_2659521
Anti-human CD45 APC/cy7 Fisher Scientific Cat#BDB557833; RRID:AB_396891
Anti-human CD279 BV421 BD Bioscience Cat#564323; AB_2738745
Anti-human CD366 APC BD Bioscience Cat#345011; RRID:AB_2561717
Anti-human CD223 PE BD Bioscience Cat#565617; RRID:AB_2889327
Anti-human CD107a PE Biolegend Cat#328608; RRID:AB_1186040
Anti-human CD69 APC Biolegend Cat#310910; RRID:AB_314845
Anti-human CD33 APC Biolegend Cat#303407; RRID:AB_314351
Anti-human EZH1 Abcam Cat#ab64850; RRID:AB_1140587
Anti-human TBP Cell Signaling Technology Cat#8515; RRID:AB_10949159
Bacterial and virus strains
Anti-CD19 CAR lentivirus Maus lab N/A
Biological samples
Mouse embryonic fibroblast feeder cells Thermo Fisher Scientific Cat#a34181
Chemicals, peptides, and recombinant proteins
DMEM/F12 STEMCELL Technologies Cat#36254
FBS Thermo Fisher Scientific Cat#353046
KnockOut serum replacement Thermo Fisher Scientific Cat#10828028
StemSpan SFEMII STEMCELL Technologies Cat##09605
StemPro-34 SFM Thermo Fisher Scientific Cat#10639011
StemFlex media Thermo Fisher Scientific Cat#A3349401
BIT serum substitute STEMCELL Technologies Cat#09500
Gentle cell dissociation reagent STEMCELL Technologies Cat#100-0485
rhVCAM-1 R&D Systems Cat#862-VC-100
SB431542 STEMCELL Technologies Cat#72234
CHIR99021 STEMCELL Technologies Cat#72054
bFGF Gemini Cat#300-112p
VEGF R&D Systems Cat#293-VE-500
BMP4 R&D Systems Cat#314-BP-500
IL-6 Peprotech Cat#200-06
IGF-1 Peprotech Cat#100-11
SCF R&D Systems Cat#255-SC-200
EPO Life Technologies Cat#PHC9634
(Continued on next page)
Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022 e1

ll
Article
Continued
FLT3 R&D Systems Cat#308-FK-250
TPO Peprotech Cat#300-18
rhDLL1/DLL4-Fc Invitrogen Cat#A42510
IL-7 R&D Systems Cat#207-IL-010
IL-15 STEMCELL Technologies Cat#78031
ImmunoCult human CD3/CD28 T cell STEMCELL Technologies Cat#10971
activator
VivoGlo Luciferin Promega Cat#P1041
Critical commercial assays
Embryoid Body Dissociation Kit Miltenyi Biotec Cat#130-096-348
CD34 MicroBead Kit Miltenyi Biotec Cat#130-046-702
LEGENDplex Hu Th1/Th2 Panel BioLegend Cat#741030
Bright-Glo Luciferase Assay System Promega Cat#E2610
Maxima cDNA Sythesis Kit Thermo Fisher Scientific Cat#FERK1641
RNeasy Plus Micro Kit QIAGEN Cat# 74034
Power SYBR Green PCR Master Mix Thermo Fisher Scientific Cat# 4367659
DNeasy Blood and Tissue Kit Qiagen Cat# 69504
Human TCRB ImmunoSEQ Assay Adaptive Biotechnologies N/A
Deposited data
Raw and processed RNA-seq data This paper GEO: GSE195667
Experimental models: Cell lines
Human 1157.2 iPSCs Daley lab N/A
Human 273 iPSCs BCH Stem Cell Core N/A
Human 1381.3 iPSCs BCH Stem Cell Core N/A
Jeko-1 cells Maus lab N/A
OCI-Ly1 cells Daley lab N/A
Experimental models: Organisms/strains
NOD-scid IL2Rgammanull (NSG) mice Jackson Laboratories Cat# 005557
Oligonucleotides
EZH1 CRISPRi gRNA: GGTGAGTGAG- Broad Institute Addgene#92385
TAAACAAGCC
DNTT forward primer: ORIGENE Cat#HP233227
5’-CAGAGCGTTCCTCATGGAGCTG-3’
DNTT reverse primer: ORIGENE Cat#HP233227
5’-GTGCTTGAAGCCACTCCAGAAC-3’
EZH1 forward primer: Integrated DNA Technologies N/A
5’-CACCACATAGTCAGTGCTTCCTG-3’
EZH1 reverse primer: Integrated DNA Technologies N/A
5’-AGTCTGACAGCGAGAGTTAGCC-3’
EZH2 forward primer: 5’- ORIGENE Cat#HP207764
GACCTCTGTCTTACTTGTGGAGC-3’
EZH2 reverse primer: 5’- ORIGENE Cat#HP207764
CGTCAGATGGTGCCAGCAATAG-3’
GAPDH forward primer: Integrated DNA Technologies N/A
5’-ACCCAGAAGACTGTGGATGG-3’
GAPDH reverse primer: Integrated DNA Technologies N/A
5’-TTCAGCTCAGGGATGACCTT-3’
Recombinant DNA
CROPseq-EZH1gRNA-Zeo plasmid This paper N/A
PB03-NDI-dCAS9-KRAB plasmid This paper N/A
(Continued on next page)
e2 Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022

ll
Article
Continued
Software and algorithms
immunoSEQ Analyzer 3.0 Adaptive Biotechnologies https://www.immunoseq.com/analyzer/
R R https://www.r-project.org/
CellRouter Lummertz da Rocha et al. (2018) https://github.com/edroaldo/cellrouter
R package FUSCA Lummertz da Rocha et al. (2022) https://github.com/edroaldo/fusca
Symphony Kang et al. (2021) https://github.com/immunogenomics/
symphony
GraphPad Prism v8 Graphpad Software, Inc https://www.graphpad.com/
FlowJo 10.8.0 BD Bioscience https://www.flowjo.com/
GSEA Broad Institute https://www.gsea-msigdb.org/gsea/
index.jsp
Aura imaging software Spectral Instruments Imaging https://spectralinvivo.com/aura-imaging-
software/
BioRender BioRender https://biorender.com/
RESOURCE AVAILABILITY
Lead contact
Further information and requests for resources and reagents should be directed to and will be fulfilled by the lead contact, George Q.
Daley ([email protected]).
Materials availability
Plasmids/reagents generated in this study will be made available upon request.
Data and code availability

d Single-cell and bulk RNA-seq data have been deposited at GEO and are publicly available as of the date of publication. Acces-
sion numbers are listed in the key resources table.
d This paper does not report original code.
d Any additional information required to reanalyze the data reported in this paper is available from the lead contact upon request.
EXPERIMENTAL MODEL AND SUBJECT DETAILS
Cell line
Human 1157.2 iPSCs established by the Stem Cell Core at Boston Children’s Hospital were maintained on Matrigel matrix (Corning,
354277) using Stemflex media (Gibco, A3349401) in 5% CO2 at 37 C.
In vivo tumor xenograft model

NOD-scid IL2Rgammanull (NSG) mice (Jackson Laboratories, 005557) were housed at the Boston Children’s Hospital animal care
facility following institutional guidelines. 8 to 12-week-old male and female mice were intravenously injected with 1x106 OCI-Ly1
DLBCL tumor cells expressing green firefly luciferase. The inoculated animals were subjected to bioluminescence imaging (BLI) using
the IVIS 200 system (PerkinElmer) twice per week following intraperitoneal injections of Vivoglo luciferin (Promega, P1043) at 150mg/
kg body weight. After two weeks, animals with substantial tumor cell engraftment (Mean total flux > 5x105 photons/sec) were
randomly assigned into four groups and intravenously injected with PBS (untreated) or 2x106 CAR T cells generated from control
iPSC-SF-T, iPSC-EZ-T, or PBMC-T cells. Tumor burden was measured by BLI weekly, and images were processed and analyzed
using Aura imaging software (Spectral Instruments Imaging). To monitor CAR T cell persistence, peripheral blood cells were collected
via retro-orbital bleeding after 3 weeks of T cell injections, and absolute numbers of CAR T cells were determined by flow cytometry
analysis. All animal experiments were performed under protocols approved by the Institutional Animal Care and Use Committee of
Boston Children’s Hospital.
METHOD DETAILS
Culture of iPSCs and generation of CD34+ HE

Human 1157.2 iPSCs were maintained on Matrigel-coated plates using Stemflex media (Gibco, A3349401) and transferred to mouse
embryonic fibroblast (MEF) feeder cells (Gibico, A34181) prior to differentiation. After one week of culture of on MEFs, iPSCs were

ll
Article
collected for the generation CD34+ HE cells using a previous described protocol (Ditadi et al., 2015). Briefly, iPSC colonies were
scraped and transferred to ultra-low attachment plates and cultured with aggregation media contains BMP4 (10ng/ml, day 0-2),
bFGF (5ng/ml, day 1-2), CHIR99021 (3mM, day 2), and SB431542 (6mM, day2) to allow EB formation. From day 3, EBs were cultured
with StemPro-34 media (Gibco, 10639011) supplemented with VEGF (15ng/ml, day3-8), bFGF (5ng/ml, day3-8), SCF (50ng/ml, day
6-8), EPO (day6-8), IL-6 (day6-8), IL-11 (day6-8), and IGF-1 (day6-8). After 8 days of culture, EBs were dissociated using EB Disso-
ciation Kit (Miltenyi, 130096348), and HE cells were isolated by magnetic-activated cell sorting (MSCS) using the human CD34 Mi-
crobead Kit (Miltenyi, 130046702).
Differentiation of iPSCs into T cells

24 well non-tissue culture treated plates were coated with 10mg/ml recombinant human DLL4-Fc protein (Life Technologies, A42510)
and 2 mg/ml VCAM-1 (Shukla et al., 2017) (R&D, 862VC100) in PBS for 2 hours at 4 C. CD34+ HE cells (1X105 cells/well) were then
seeded on DLL4-coated plates and cultured in SFEMII (StemCell Tech, 09605) media supplemented with 10% BIT serum substitute
(StemCell Tech, 09500), 2mM L-Glutamine (Gibco, 35050061), 1mM sodium pyruvate (Gibco, 11360070), 50mg/ml 2-phospho-L-
ascorbic acid (Huijskens et al., 2014) (sigma, 49752), 55mM 2-Mercaptoethanol (Gibco, 21985023), 1mM None-essential amino acids
(Gibco, 11140050), 1% penicillin-streptomycin (Corning, 30002CI), 30ng/ml SCF, 20ng/ml FLT3, 30ng/ml IL-7, 5ng/ml IL-3 (first week
only), and 5ng/ml TPO for 2 weeks to induced the differentiation into proT cells. For the production of EZ-T cells, EZH1 knockdown
was performed via lentiviral transduction (MOI=5) after 24 hours of HE seeding. After 2 weeks of differentiation, SCF and TPO were
withdrawn from the media, and the proT cells were replated into new DLL4-coated plates and cultured in the presence of 20ng/ml
FLT3 and 15ng/ml IL-7 till week 5, followed by 1 week of treatment with CD3/CD28 T activator (StemCell Tech, 10971) and 5ng/ml IL-
15 to induced SP T cells. During T cell differentiation, half media changes were conducted every 2-3 days. T cell differentiation using
the OP9-DL1 co-culture system was conducted following previous reports (Holmes and Zúñiga-Pflu €cker, 2009; Themeli et al., 2013).
Briefly, day 8 CD34+ HE cells were seeded on OP-DL1 stromal cells and co-cultured with OP9 media (a-MEME, 20% FBS, 1% peni-
cillin-streptomycin, 1mM None-essential amino acids, 2mM L-Glutamine, 10mM 2-Mercaptoethanol, 50mg/ml 2-phospho-L-ascorbic
acid) supplemented with 10ng/ml SCF, 5ng/ml FLT3, and 10ng/ml IL-7. Cells were mechanically dissociated and filtered through
40mm strainers to be seeded on new stromal cells every five days.
CRISPR interference in iPSs

gRNA targeting the TSS of EZH1 was pulled from the Broad Institute’s genome-wide Dolcetto library (Addgene #92385, Library Set A)
(Sanson et al., 2018) and ordered as oligos from IDT. gRNA oligos (GGTGAGTGAGTAAACAAGCC) were annealed, phosphorylated
and cloned by Golden Gate Assembly with BsmBI into a modified CROPseq-Zeo vector constitutively expressing mNeon and a
modified gRNA scaffold sequence as previously described (Dang et al., 2015). Lentiviruses for each gRNA vector was generated
by transfection of pMD2.G (Addgene #12259), psPAX2 (Addgene #12260), and the successfully cloned CROPseq transfer plasmid
(2:3:4 ratio by mass and 3ug total plasmid) into HEK293FT cells using Lipofectamine 3000 (Thermo Fisher L3000015). Viral superna-
tant was harvested 48 hours after transfection and filtered through 0.45 mm PVDF filters (Millipore SLHVR04NL). Human 1157.2 iPSC
line expressing a doxycycline-inducible dCas9-KRAB-2A-mCherry cassette (from Boston Children’s Hospital Stem Cell Core) were
singularized with TrypLE, plated in a matrigel-coated 6-well plate at 165,000 cells per well, and then infected with individual CROPseq
lentiviruses at MOI=0.25 with 8ug/mL polybrene. The following day the media was replenished and the cells were selected with
200 mg/mL zeocin (ThermoFisher R25001) for 24 hours. Selected hiPSCs were scaled up, banked, and used for downstream
differentiation.
Immunoblot
Whole cell lysates were collected using NP-40 lysis buffer (Invitrogen, FNN0021) with protease and phosphatase inhibitor cocktail
(Thermo Scientific, 78443). 50mg total protein was separated by SDS–PAGE using Any kDTM Mini-protean TGX precast polyacryl-
amide gels (BioRad, 4569033), and transferred to PVDF membranes using the Trans-Blot Cell Transfer System (BioRad). Membranes
were incubated overnight with antibodies against TBP (Cell Signaling, 8515, 1:1000), EZH1 (Abcam, ab64850, 1:1000) at 4 C, fol-
lowed by incubation with horseradish peroxidase-conjugated secondary antibodies (1:2000) for 1 hour at room temperature. Chem-
iluminescence was detected using SuperSignal West Pico Plus Chemiluminescent substrate (Thermo Scientific, 34579).
Quantitative Real-Time PCR analysis

RNA extraction and removal of genomic DNA was performed using the RNeasy Mini Kit (Qiagen, 74104). First strand cDNA was syn-
thesized using Maxiam First Strand cDNA SynthesisKit (Thermo Scientific, K1641). Quantitative real-time PCR was performed on a
QuantStudio 7 Flex Real-Time PCR machine (Applied Biosystems, 4485701) using Power SYBR Green PCR Master Mix (Applied Bio-
systems, 4367659) following the manufacturer’s directions. The following oligonucleotides were used: DNTT forward primer: 5’-CA
GAGCGTTCCTCATGGAGCTG-3’; DNTT reverse primer: 5’-GTGCTTGAAGCCACTCCAGAAC-3’; EZH1 forward primer: 5’-CACCA
CATAGTCAGTGCTTCCTG-3’; EZH1 reverse primer: 5’-AGTCTGACAGCGAGAGTTAGCC-3’; GAPDH forward primer: 5’-ACCCA
GAAGACTGTGGATGG-3’; GAPDH reverse primer: 5’-TTCAGCTCAGGGATGACCTT-3’; EZH2 forward primer: 5’- GACCCTG
TCTTACTTGTGGAGC-3’; EZH2 reverse primer: 5’- CGTCAGATGGTGCCAGCAATAG-3’.

ll
Article
Flow cytometry
Cells were stained with PI/DAPI (BD Biosciences, RUO) and antibodies at 1:100 dilution in PBS with 2% FBS for 30 min at room
temperature in the dark. BD LSRII and BD FACSAria II were used for flow cytometry analysis and cells sorting. Compensation
was performed by automated compensation with anti-mouse Igk and negative beads (BD Biosciences) using the BD FACSDiva soft-
ware. The following human antibodies were used: CD45RA-BV510 (BD Biosciences, H100), CD45RO-PE (BD Biosciences, UCHL1),
CCR7-APC (BD Biosciences, 2-L1-A), CD3-PE/Cy7 (Biolegend, SK7), CD8a-BV421 (BD Biosciences, RPA-T8), TCRab-APC (Bio-
legend, IP26), TCRgd-PE (Biolegend, B1), CD7-PE (BD Biosciences, M-T701), CD5-BV510 (BD Biosciences, UCHT2), CD4-PE/
Cy5 (Biolegend, RPA-T4), CD8b-APC(Miltenyi, REA715), CD45-APC/Cy7 (BD Biosciences, 2D1), CD279-BV421 (BD Biosciences,
EH12.1), CD366-APC (BD Biosciences, F38-2E2), CD223-PE (BD Biosciences, T47-530), CD107a-PE (Biolegend, LAMP-1), CD69
APC (Biolegend, FN50), CD33-APC (Biolegend, WM53).
Bulk RNA-seq and data analysis

Total RNA samples were isolated from iPSC-derived CD3+ cells or PBMC NK/T cells using a column assay with the DNase treatment
(Direct-zol MicroPrep, ZYMO). Quantity and quality of the RNA were evaluated using the nanodrop machine and RNA screen tape.
High quality RNA (Both 280/260 and 230/260 over 1.7 with RNA integrity number >7) underwent ribosomal RNA depletion and then
library construction. For regular gene expression analysis, adaptor trimmed reads from the sequencer were mapped to the human
genome, quantified, and analyzed using seq data analysis tools (Cutadapt, Bowtie, TopHat, HTSeq, R, and edgeR). The RNA-seq
data is available in GEO database (GSE195667). Portions of this research were conducted on the O2 High Performance Compute
Cluster, supported by the Research Computing Group, at Harvard Medical School.
TCR repertoire analysis

CD3+ T cells were FACS-isolated from control PBMC T cells or iPSC-derived T cells, followed by gDNA extraction using the Qiagen
DNeasy Blood & Tissue Kit (Qiagen, 69504). DNA samples were then subjected to TCRB sequencing via immunoSEQ assay. (Adap-
tive Biotechnologies). TCR rearrangements, Vbgene usage, and CDR3 length were analyzed using the immunoSEQ Analyzer 3.0 soft-
ware (Adaptive Biotechnologies).
Single-cell RNA-Sequencing and data analysis

T cells were FACS-isolated for DAPI-CD45+ expression following in vitro differentiation and/or activation. Single-cell suspensions
were loaded onto a Chromium Single Cell Chip (10X Genomics) according to the manufacturer’s instructions for a target recovery
of 6000 cells per lane. Each sample was loaded into two lanes to serve as technical replicates for scRNA-Seq library
preparation. Libraries were then prepared per the 10X scRNA-Seq v2 protocol in parallel for all conditions. Final 10X scRNA-Seq li-
braries were assayed via an Agilent High Sensitivity dsDNA Bioanalyzer, normalized, pooled and shallow sequenced on a MiniSeq to
quantify the number of high confidence cell barcodes. The libraries were then renormalized per the distribution of reads/library from
the MiniSeq run and deep sequenced on a NextSeq to a depth of 50,000 reads per cell barcode.
Sequencing libraries were computationally demultiplexed and the data were aligned to the GRCh38 reference genome using cell-
ranger v2.1 (10X Genomics). CellRouter was used for quality control and downstream analysis (Lummertz da Rocha et al., 2018). Cell
by gene count matrices of all samples were combined to a single gene expression matrix. Cells with 500–3000 detected genes and
expressing <10% mitochondrial genes as well as genes expressed in > 25 cells were retained for downstream analysis. The final
dataset after quality control was composed of 10907 cells, with a median of 1,456 genes detected per cell. Variation caused by
mitochondrial gene expression and sample replicates were regressed out. All genes passing QC metrics were used for principal
component analysis. Clustering analysis was performed using parameters k=150 and 30 principal components. UMAP analysis
was performed using 10 principal components. Cluster annotation was performed using differential expression and marker genes.
Gene signatures, as well as ranked gene lists ordered by log fold change for GSEA analysis, were generated by testing for differential
expression of a cluster/cell type against all other cells using a Wilcox test, as implemented in CellRouter. Trajectory analysis and
calculation of GRN scores was performed with CellRouter. SCENIC analysis was used to identify regulons enriched in each anno-
tated cell type (Aibar et al., 2017). Data generated by Granja et al. (2019) were downloaded from https://github.com/
GreenleafLab/MPAL-Single-Cell-2019 and converted into a FUSCA object (Lummertz da Rocha et al., 2022). We used FUSCA to
perform UMAP analysis and prepare input parameters for reference mapping using Symphony (Kang et al., 2021).
T cell activation and degranulation

Control T cells generated from primary PBMC T cells of heathy donors and iPSC-T cells were treated with T cell stimulation cocktail
(Invitrogen, 00497093) containing phorbol 12-myristate 13-acetate (PMA) and ionomycin for 6 hours. Percentage of iPSC cells that
express CD3 and CD69 was measured by flow cytometry to determine T cell activation. Similarly, percentage of CD8+CD3+ iPSC
cells that express CD107a was measured by flow cytometry to detect T cell degranulation.
CAR T cell functional assays

Control PBMC T cells and iPSC-derived T cells were activated (day 0) using anti- CD3/CD28 Dynabeads (Gibco, 11131D) or CD3/
CD28 T activator (StemCell Tech, 10971), followed by transduction with a lentiviral vector encoding the CD19-CAR 24-hours later
(Scarfò et al., 2018). T cells were cultured in RPMI media containing 10% fetal bovine serum, penicillin, streptomycin and

ll
Article
supplemented with 20 IU/ml rhIL-2 beginning on day 0 of culture and were maintained at a constant cell concentration of 0.5 x106/mL
by counting every 2-3 days. On day 10 cells were de-beaded (when using Dynabeads) and used for assays. For cytotoxicity assays,
control PBMC or iPSC-T cells were co-cultured with luciferase-expressing Jeko-1 or OCI-Ly1 tumor cells at the indicated ratios for 18
hours. Luciferase activity was measured with a Synergy Neo2 luminescence microplate reader (Biotek). Percentage of specific lysis
was calculated as (total RLU / target cells only RLU) x100. Cell-free supernatants were collected for cytokine release assay. Levels of
cytokines were measured using a LEGENDplex Multiplex Assay Kit (Biolegend, 741030) following manufacture’s instructions.
QUANTIFICATION AND STATISTICAL ANALYSIS
The significance of the difference between control and experimental results generated from in vitro assays was determined by two
tailed student’s T test, and P < 0.05 was considered statistically significant. For in vivo experiments, survival results were presented in
a Kaplan-Meier survival plot and compared by log-rank (Mantel-Cox) test. Statistic parameters in each experiment (value of n, SEM,
or SD) are described in the figure legends.

EZH1 iPSC CAR T

Uploaded by

Copyright:

Available Formats

EZH1 iPSC CAR T

Uploaded by

Document Information

Copyright

Available Formats

Share this document

Share or Embed Document

Sharing Options

Did you find this document useful?

Is this content inappropriate?

Copyright:

Available Formats

EZH1 iPSC CAR T

Uploaded by

Copyright:

Available Formats

Article

EZH1 repression generates mature iPSC-derived

d EZH1 repression generates mature EZ-T cells similar to pe-

d EZ-T cells can give rise to memory-like T cells upon activation

d CAR EZ-T cells display enhanced antitumor activity in vitro

Jing et al., 2022, Cell Stem Cell 29, 1181–1196

INTRODUCTION bor-intensive, and expensive therapy is still not widely available.

(legend on next page)

1182 Cell Stem Cell 29, 1181–1196, August 4, 2022

Figure 1. Stroma-free differentiation of human iPSCs into T cells

Cell Stem Cell 29, 1181–1196, August 4, 2022 1183

1184 Cell Stem Cell 29, 1181–1196, August 4, 2022

Figure 2. EZH1 knockdown facilitates in vitro T cell differentiation from iPSCs

Cell Stem Cell 29, 1181–1196, August 4, 2022 1185

Figure 3. EZ-T cells display molecular features of mature TCRab T cells

1186 Cell Stem Cell 29, 1181–1196, August 4, 2022

Figure 4. TCR repertoire analysis of EZ-T cells

Cell Stem Cell 29, 1181–1196, August 4, 2022 1187

1188 Cell Stem Cell 29, 1181–1196, August 4, 2022

Cell Stem Cell 29, 1181–1196, August 4, 2022 1189

Figure 6. EZ-T cells display enhanced effector functions

1190 Cell Stem Cell 29, 1181–1196, August 4, 2022

(legend on next page)

Cell Stem Cell 29, 1181–1196, August 4, 2022 1191

1192 Cell Stem Cell 29, 1181–1196, August 4, 2022

outcomes in adoptive T cell therapies (Gattinoni et al., 2005; Kle- STAR+METHODS

graft-versus-host disease (GVHD) inherent in allogeneic T cell B Materials availability

overcome the suppressive tumor microenvironment (Pegram B TCR repertoire analysis

Cell Stem Cell 29, 1181–1196, August 4, 2022 1193

1194 Cell Stem Cell 29, 1181–1196, August 4, 2022

Cell Stem Cell 29, 1181–1196, August 4, 2022 1195

1196 Cell Stem Cell 29, 1181–1196, August 4, 2022

KEY RESOURCES TABLE

REAGENT or RESOURCE SOURCE IDENTIFIER

Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022 e1

e2 Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022

Data and code availability

EXPERIMENTAL MODEL AND SUBJECT DETAILS

In vivo tumor xenograft model

Culture of iPSCs and generation of CD34+ HE

Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022 e3

Differentiation of iPSCs into T cells

CRISPR interference in iPSs

Quantitative Real-Time PCR analysis

e4 Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022

Bulk RNA-seq and data analysis

TCR repertoire analysis

Single-cell RNA-Sequencing and data analysis

T cell activation and degranulation

CAR T cell functional assays

Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022 e5

QUANTIFICATION AND STATISTICAL ANALYSIS

e6 Cell Stem Cell 29, 1181–1196.e1–e6, August 4, 2022

You might also like