Il-2 delivery by engineered mesenchymal stem cells re-invigorates cd8+ t cells to overcome immunotherapy resistance in cancer

ABSTRACT Immune checkpoint blockade (ICB)-based immunotherapy depends on functional tumour-infiltrating lymphocytes (TILs), but essential cytokines are less understood. Here we uncover an

essential role of endogenous IL-2 for ICB responsiveness and the correlation between insufficient IL-2 signalling and T-cell exhaustion as tumours progress. To determine if exogenous IL-2 in

the tumour microenvironment can overcome ICB resistance, we engineered mesenchymal stem cells (MSCs) to successfully deliver IL-2 mutein dimer (SIL2-EMSC) to TILs. While MSCs have been used

to suppress inflammation, SIL2-EMSCs elicit anti-tumour immunity and overcome ICB resistance without toxicity. Mechanistically, SIL2-EMSCs activate and expand pre-existing CD8+ TILs,

sufficient for tumour control and induction of systemic anti-tumour effects. Furthermore, engineered MSCs create synergy of innate and adaptive immunity. The therapeutic benefits of

SIL2-EMSCs were also observed in humanized mouse models. Overall, engineered MSCs rejuvenate CD8+ TILs and thus potentiate ICB and chemotherapy. Access through your institution Buy or

subscribe This is a preview of subscription content, access via your institution ACCESS OPTIONS Access through your institution Access Nature and 54 other Nature Portfolio journals Get

Nature+, our best-value online-access subscription $32.99 / 30 days cancel any time Learn more Subscribe to this journal Receive 12 print issues and online access $209.00 per year only

$17.42 per issue Learn more Buy this article * Purchase on SpringerLink * Instant access to full article PDF Buy now Prices may be subject to local taxes which are calculated during checkout

ADDITIONAL ACCESS OPTIONS: * Log in * Learn about institutional subscriptions * Read our FAQs * Contact customer support SIMILAR CONTENT BEING VIEWED BY OTHERS MESENCHYMAL STROMAL CELLS

EQUIPPED BY IFNΑ EMPOWER T CELLS WITH POTENT ANTI-TUMOR IMMUNITY Article Open access 10 February 2022 TUMOR-CONDITIONAL IL-15 PRO-CYTOKINE REACTIVATES ANTI-TUMOR IMMUNITY WITH LIMITED

TOXICITY Article 10 August 2021 ANTI-PD-1 CIS-DELIVERY OF LOW-AFFINITY IL-12 ACTIVATES INTRATUMORAL CD8+T CELLS FOR SYSTEMIC ANTITUMOR RESPONSES Article Open access 03 June 2024 DATA

AVAILABILITY scRNA-seq data that support the findings of this study (Fig. 1a–d and Extended Data Fig. 1) can be accessed through the Gene Expression Omnibus under accession code GSE178881.

The human SKCM data were derived from the TCGA Research Network: http://cancergenome.nih.gov/. Cumulative survival rate in patients with SKCM and gene correlation were analysed using Tumor

Immune Estimation Resource (TIMER; https://cistrome.shinyapps.io/timer/). Source data are provided with this paper, available online for Figs. 1–7 and Extended Data Figs. 1–10. All other

data that support the findings of this study are available from the corresponding author on reasonable request. CODE AVAILABILITY The scRNA data were processed using Cell Ranger v.2.1.1

(https://www.10xgenomics.com/) and analysed with the R package Seurat v.3.1.2 (https://satijalab.org/seurat/). The R packages fgesa v.1.16.0

(http://bioconductor.org/packages/release/bioc/html/fgsea.html) and msigdbr v.7.2.1 (https://cran.r-project.org/web/packages/msigdbr/index.html) were used to perform the GSEA. REFERENCES *

Jiang, Y., Li, Y. & Zhu, B. T-cell exhaustion in the tumor microenvironment. _Cell Death Dis._ 6, e1792 (2015). Article  CAS  Google Scholar  * Baitsch, L., Fuertes-Marraco, S. A.,

Legat, A., Meyer, C. & Speiser, D. E. The three main stumbling blocks for anticancer T cells. _Trends Immunol._ 33, 364–372 (2012). Article  CAS  Google Scholar  * Thommen, D. S. &

Schumacher, T. N. T cell dysfunction in cancer. _Cancer Cell_ 33, 547–562 (2018). Article  CAS  Google Scholar  * Im, S. J. et al. Defining CD8+ T cells that provide the proliferative burst

after PD-1 therapy. _Nature_ 537, 417–421 (2016). Article  CAS  Google Scholar  * Jansen, C. S. et al. An intra-tumoral niche maintains and differentiates stem-like CD8 T cells. _Nature_

576, 465–470 (2019). Article  CAS  Google Scholar  * Miller, B. C. et al. Subsets of exhausted CD8+ T cells differentially mediate tumor control and respond to checkpoint blockade. _Nat.

Immunol._ 20, 326–336 (2019). Article  CAS  Google Scholar  * Siddiqui, I. et al. Intratumoral Tcf1+PD-1+CD8+ T cells with stem-like properties promote tumor control in response to

vaccination and checkpoint blockade immunotherapy. _Immunity_ 50, 195–211 e110 (2019). Article  CAS  Google Scholar  * Kurtulus, S. et al. Checkpoint blockade immunotherapy induces dynamic

changes in PD-1−CD8+ tumor-infiltrating T cells. _Immunity_ 50, 181–194 e186 (2019). Article  CAS  Google Scholar  * Jenkins, R. W., Barbie, D. A. & Flaherty, K. T. Mechanisms of

resistance to immune checkpoint inhibitors. _Br. J. Cancer_ 118, 9–16 (2018). Article  CAS  Google Scholar  * Ayers, M. et al. IFN-γ-related mRNA profile predicts clinical response to PD-1

blockade. _J. Clin. Invest._ 127, 2930–2940 (2017). Article  Google Scholar  * Rosenberg, S. A. IL-2: the first effective immunotherapy for human cancer. _J. Immunol._ 192, 5451–5458 (2014).

Article  CAS  Google Scholar  * Schwager, K., Hemmerle, T., Aebischer, D. & Neri, D. The immunocytokine L19-IL2 eradicates cancer when used in combination with CTLA-4 blockade or with

L19-TNF. _J. Invest. Dermatol._ 133, 751–758 (2013). Article  CAS  Google Scholar  * Buchbinder, E. I. et al. Therapy with high-dose interleukin-2 (HD IL-2) in metastatic melanoma and renal

cell carcinoma following PD1 or PDL1 inhibition. _J. Immunother. Cancer_ 7, 49 (2019). Article  Google Scholar  * Pauken, K. E. & Wherry, E. J. Overcoming T cell exhaustion in infection

and cancer. _Trends Immunol._ 36, 265–276 (2015). Article  CAS  Google Scholar  * Zhang, L. et al. Lineage tracking reveals dynamic relationships of T cells in colorectal cancer. _Nature_

564, 268–272 (2018). Article  CAS  Google Scholar  * Gubin, M. M. et al. High-dimensional analysis delineates myeloid and lymphoid compartment remodeling during successful immune-checkpoint

cancer therapy. _Cell_ 175, 1014–1030 e1019 (2018). Article  CAS  Google Scholar  * Khan, O. et al. TOX transcriptionally and epigenetically programs CD8+ T cell exhaustion. _Nature_ 571,

211–218 (2019). Article  CAS  Google Scholar  * Alfei, F. et al. TOX reinforces the phenotype and longevity of exhausted T cells in chronic viral infection. _Nature_ 571, 265–269 (2019).

Article  CAS  Google Scholar  * Ren, Z. et al. Selective delivery of low-affinity IL-2 to PD-1+ T cells rejuvenates antitumor immunity with reduced toxicity. _J. Clin. Invest._ 132, e153604

(2022). Article  CAS  Google Scholar  * Sakuishi, K. et al. Targeting Tim-3 and PD-1 pathways to reverse T cell exhaustion and restore anti-tumor immunity. _J. Exp. Med._ 207, 2187–2194

(2010). Article  CAS  Google Scholar  * Sun, Z. et al. A next-generation tumor-targeting IL-2 preferentially promotes tumor-infiltrating CD8+ T-cell response and effective tumor control.

_Nat. Commun._ 10, 3874 (2019). Article  CAS  Google Scholar  * Mott, H. R. et al. The solution structure of the F42A mutant of human interleukin 2. _J. Mol. Biol._ 247, 979–994 (1995).

Article  CAS  Google Scholar  * Levin, A. M. et al. Exploiting a natural conformational switch to engineer an interleukin-2 ‘superkine’. _Nature_ 484, 529–533 (2012). Article  CAS  Google

Scholar  * Schlothauer, T. et al. Novel human IgG1 and IgG4 Fc-engineered antibodies with completely abolished immune effector functions. _Protein Eng. Des. Sel._ 29, 457–466 (2016). Article

  CAS  Google Scholar  * Liu, L. et al. Rejuvenation of tumour-specific T cells through bispecific antibodies targeting PD-L1 on dendritic cells. _Nat. Biomed. Eng._ 5, 1261–1273 (2021).

Article  CAS  Google Scholar  * Hsu, E. J. et al. A cytokine receptor-masked IL2 prodrug selectively activates tumor-infiltrating lymphocytes for potent antitumor therapy. _Nat. Commun._ 12,

2768 (2021). Article  CAS  Google Scholar  * Kidd, S. et al. Direct evidence of mesenchymal stem cell tropism for tumor and wounding microenvironments using in vivo bioluminescent imaging.

_Stem Cells_ 27, 2614–2623 (2009). Article  CAS  Google Scholar  * Chulpanova, D. S. et al. Application of mesenchymal stem cells for therapeutic agent delivery in anti-tumor treatment.

_Front. Pharm._ 9, 259 (2018). Article  Google Scholar  * Zou, W. et al. LIGHT delivery to tumors by mesenchymal stem cells mobilizes an effective antitumor immune response. _Cancer Res._

72, 2980–2989 (2012). Article  CAS  Google Scholar  * Lv, F. J., Tuan, R. S., Cheung, K. M. & Leung, V. Y. Concise review: the surface markers and identity of human mesenchymal stem

cells. _Stem Cells_ 32, 1408–1419 (2014). Article  CAS  Google Scholar  * Peister, A. et al. Adult stem cells from bone marrow (MSCs) isolated from different strains of inbred mice vary in

surface epitopes, rates of proliferation, and differentiation potential. _Blood_ 103, 1662–1668 (2004). Article  CAS  Google Scholar  * van Deursen, J. M. The role of senescent cells in

ageing. _Nature_ 509, 439–446 (2014). Article  Google Scholar  * Turinetto, V., Vitale, E. & Giachino, C. Senescence in human mesenchymal stem cells: functional changes and implications

in stem cell-based therapy. _Int. J. Mol. Sci._ 17, 1164 (2016). Article  Google Scholar  * Bonab, M. M. et al. Aging of mesenchymal stem cell in vitro. _BMC Cell Biol._ 7, 14 (2006).

Article  Google Scholar  * Westerman, K. A. & Leboulch, P. Reversible immortalization of mammalian cells mediated by retroviral transfer and site-specific recombination. _Proc. Natl

Acad. Sci. USA_ 93, 8971–8976 (1996). Article  CAS  Google Scholar  * Iida, Y. et al. Local injection of CCL19-expressing mesenchymal stem cells augments the therapeutic efficacy of

anti-PD-L1 antibody by promoting infiltration of immune cells. _J. Immunother. Cancer_ 8, e000582 (2020). Article  Google Scholar  * Huang, X. et al. Leveraging an NQO1 bioactivatable drug

for tumor-selective use of poly(ADP-ribose) polymerase inhibitors. _Cancer Cell_ 30, 940–952 (2016). Article  CAS  Google Scholar  * Li, X. et al. NQO1 targeting prodrug triggers innate

sensing to overcome checkpoint blockade resistance. _Nat. Commun._ 10, 3251 (2019). Article  Google Scholar  * Thompson, E. D., Enriquez, H. L., Fu, Y. X. & Engelhard, V. H. Tumor masses

support naive T cell infiltration, activation, and differentiation into effectors. _J. Exp. Med._ 207, 1791–1804 (2010). Article  CAS  Google Scholar  * Gupta, P. K. et al. CD39 expression

identifies terminally exhausted CD8+ T cells. _PLoS Pathog._ 11, e1005177 (2015). Article  Google Scholar  * Chen, Z. et al. TCF-1-centered transcriptional network drives an effector versus

exhausted CD8 T cell-fate decision. _Immunity_ 51, 840–855 e845 (2019). Article  CAS  Google Scholar  * Guo, Y. et al. Metabolic reprogramming of terminally exhausted CD8+ T cells by IL-10

enhances anti-tumor immunity. _Nat. Immunol._ 22, 746–756 (2021). Article  CAS  Google Scholar  * Krishna, S. et al. Stem-like CD8 T cells mediate response of adoptive cell immunotherapy

against human cancer. _Science_ 370, 1328–1334 (2020). Article  CAS  Google Scholar  * Liu, J. et al. Improved efficacy of neoadjuvant compared to adjuvant immunotherapy to eradicate

metastatic disease. _Cancer Discov._ 6, 1382–1399 (2016). Article  CAS  Google Scholar  * Forde, P. M. et al. Neoadjuvant PD-1 blockade in resectable lung cancer. _N. Engl. J. Med._ 378,

1976–1986 (2018). Article  CAS  Google Scholar  * Mender, I. et al. Telomere stress potentiates STING-dependent anti-tumor immunity. _Cancer Cell_ 38, 400–411 e406 (2020). Article  CAS 

Google Scholar  * Arai, S. et al. Infusion of the allogeneic cell line NK-92 in patients with advanced renal cell cancer or melanoma: a phase I trial. _Cytotherapy_ 10, 625–632 (2008).

Article  CAS  Google Scholar  * Nowakowska, P. et al. Clinical grade manufacturing of genetically modified, CAR-expressing NK-92 cells for the treatment of ErbB2-positive malignancies.

_Cancer Immunol. Immunother._ 67, 25–38 (2018). Article  CAS  Google Scholar  * Pink, J. J. et al. NAD(P)H:quinone oxidoreductase activity is the principal determinant of β-lapachone

cytotoxicity. _J. Biol. Chem._ 275, 5416–5424 (2000). Article  CAS  Google Scholar  * Bae, J. et al. Phc2 controls hematopoietic stem and progenitor cell mobilization from bone marrow by

repressing Vcam1 expression. _Nat. Commun._ 10, 3496 (2019). Article  Google Scholar  * Semenza, G. L. et al. Hypoxia response elements in the aldolase A, enolase 1, and lactate

dehydrogenase A gene promoters contain essential binding sites for hypoxia-inducible factor 1. _J. Biol. Chem._ 271, 32529–32537 (1996). Article  CAS  Google Scholar  * Zappasodi, R. et al.

CTLA-4 blockade drives loss of Treg stability in glycolysis-low tumours. _Nature_ 591, 652–658 (2021). Article  CAS  Google Scholar  * Scrimieri, F. et al. Murine leukemia virus envelope

gp70 is a shared biomarker for the high-sensitivity quantification of murine tumor burden. _Oncoimmunology_ 2, e26889 (2013). Article  Google Scholar  * Qiao, J. et al. Targeting tumors with

IL-10 prevents dendritic cell-mediated CD8+ T cell apoptosis. _Cancer Cell_ 35, 901–915 e904 (2019). Article  CAS  Google Scholar  * Liu, L. et al. Concurrent delivery of immune checkpoint

blockade modulates T cell dynamics to enhance neoantigen vaccine-generated antitumor immunity. _Nat. Cancer_ 3, 437–452 (2022). Article  CAS  Google Scholar  Download references

ACKNOWLEDGEMENTS This work was supported by Cancer Prevention and Research Institute of Texas (CPRIT) grant RR150072 given to Y.-X.F. and the NIH/NCI grant R01-CA240952 given to J.Q. The

funders had no role in study design, data collection and analysis, decision to publish or preparation of the manuscript. We thank the Institutional Animal Care and Use Committee Animal

Resources Center, and Animal Research Center. SSR#69 plasmid was kindly provided by T. C. He at Chicago University. HIF-1 reporter p2.1 plasmid was kindly provided by W. Luo at UT

Southwestern Medical Center. Human cord blood was kindly provided by R. A. Word at Obstetrics and Gynecology Tissue Procurement Facility in UT Southwestern Medical Center, supported by

NIH-P01-HD087150. We also thank C. Han, Z. Liu, C. Lu, Y. Liang, X. Cao, C. Dong and B. Moon for providing experiment materials and helpful discussions. AUTHOR INFORMATION Author notes *

These authors contributed equally to this work: Joonbeom Bae, Longchao Liu, Casey Moore. AUTHORS AND AFFILIATIONS * Department of Pathology, University of Texas Southwestern Medical Center,

Dallas, TX, USA Joonbeom Bae, Longchao Liu, Casey Moore, Eric Hsu, Anli Zhang, Zhenhua Ren, Zhichen Sun, Xue Wang, Jiankun Zhu, Jian Qiao & Yang-Xin Fu * Department of Immunology,

University of Texas Southwestern Medical Center, Dallas, TX, USA Casey Moore & Eric Hsu * Department of Pharmacology, Harold C. Simmons Comprehensive Cancer Center, University of Texas

Southwestern Medical Center, Dallas, TX, USA Zhichen Sun * Key Laboratory of Infection and Immunity, Institute of Biophysics, Chinese Academy of Sciences, Beijing, China Jiao Shen *

University of Chinese Academy of Sciences, Beijing, China Jiao Shen * Department of Basic Medical Sciences, School of Medicine, Tsinghua University, Beijing, China Yang-Xin Fu Authors *

Joonbeom Bae View author publications You can also search for this author inPubMed Google Scholar * Longchao Liu View author publications You can also search for this author inPubMed Google

Scholar * Casey Moore View author publications You can also search for this author inPubMed Google Scholar * Eric Hsu View author publications You can also search for this author inPubMed 

Google Scholar * Anli Zhang View author publications You can also search for this author inPubMed Google Scholar * Zhenhua Ren View author publications You can also search for this author

inPubMed Google Scholar * Zhichen Sun View author publications You can also search for this author inPubMed Google Scholar * Xue Wang View author publications You can also search for this

author inPubMed Google Scholar * Jiankun Zhu View author publications You can also search for this author inPubMed Google Scholar * Jiao Shen View author publications You can also search for

this author inPubMed Google Scholar * Jian Qiao View author publications You can also search for this author inPubMed Google Scholar * Yang-Xin Fu View author publications You can also

search for this author inPubMed Google Scholar CONTRIBUTIONS Conceptualization, J.B. and Y.-X.F.; methodology, J.B. and Y.-X.F.; investigation, J.B., L.L., C.M., E.H, J.S. and X.W.;

writing—original draft, J.B.; writing—review and editing, C.M., E.H., J.Q. and Y.-X.F.; funding acquisition, Y.-X.F. and J.Q.; resources, A.Z., L.L., Z.S., Z.R. and J.Z.; supervision,

Y.-X.F. CORRESPONDING AUTHORS Correspondence to Jian Qiao or Yang-Xin Fu. ETHICS DECLARATIONS COMPETING INTERESTS The authors declare no competing interests. PEER REVIEW PEER REVIEW

INFORMATION _Nature Cell Biology_ thanks Weiyi Peng, George Coukos and the other, anonymous, reviewer(s) for their contribution to the peer review of this work. ADDITIONAL INFORMATION

PUBLISHER’S NOTE Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations. EXTENDED DATA EXTENDED DATA FIG. 1 SINGLE CELL

ANALYSIS OF CD8+ T CELLS IN EARLY AND ADVANCED TUMOR. A, Schematic workflow of single cell RNA-Seq data generation. B, Feature plot showing the hallmark genes in each single-cell cluster

identified by UMAP. C, UMAP plot showing the distribution of CD8+ T cells in the spleen of day 10 (S10), tumor of day 10 (T10), spleen of day 20 (S20), tumor of day 20 (T20). D, Violin plots

showing the expression levels of IL-2 related genes across CD8+ T cell clusters. E, the Gene Set Enrichment Analysis (GSEA) was performed between cluster 1 (advanced tumor specific) and

cluster 2 (early tumor specific). Source data EXTENDED DATA FIG. 2 ADVANCED TUMORS CONTAIN REDUCED NUMBER OF CD8 + TILS AND PREDOMINANTLY TERMINALLY EXHAUSTED CELLS. A-C, C57BL/6J mice (T10,

n = 6; T20, n = 8) were s.c. inoculated with 1 × 106 MC38 cells. 10 days (T10) or 20 days (T20) after tumor inoculation, TILs were analyzed for the frequency of PD1−TIM3−, PD1 + TIM3−, and

PD1+TIM3+ subset (A), stem-like exhausted CD8+ T cells (B), terminally exhausted CD8+ T cells (C). D, C57BL/6 J mice (Control, n = 6; anti-PD-L1, n = 5; anti-PD-L1 + anti-IL15, n = 6;

anti-PD-L1 + anti-IL2, n = 5; anti-PD-L1 + anti-IL2Rβ, n = 5) were s.c. inoculated with 1 × 106 MC38 cells and i.p. injected with 150 μg of anti-PD-L1 and/or 200 μg anti-IL2Rβ, 200 μg

anti-IL2 or 200 μg anti-IL15 on day 7,10 and 13 (black arrow). Tumor growth was measured twice a week. Data are shown as mean ± s.e.m. from two independent experiments. _P_ value was

determined by two-tailed unpaired t test (A-C) or two-way ANOVA (D). Source data are available online. Source data EXTENDED DATA FIG. 3 SURVIVAL, BIODISTRIBUTION AND SIL2 PRODUCTION OF

ENGINEERED MSC. A, Functional activity of WTIL2 and SIL2 were assessed using the HEK-BlueTM IL-2 reporter cell assay (n = 2). B,C, CT26 bearing BALB/c mice (n = 4 per group) were treated

with SIL2 (20 μg, i.t.) on day 9. Five days after treatment, the concentration of SIL2 in serum (B), tumor and other tissues (C) were determined by hIgG ELISA. D, Body weight of CT26 bearing

BALB/c mice (n = 5 per group) treated with SIL2-EMSC (1 × 106, p.t.) or SIL2 (20 μg, i.t.) on day 9 and 12. E, Surface marker expression of MSC by passage number in primary MSC and

immortalized MSC (iMSC). F, Proliferation of CFSE-labeled primary MSC and iMSC assessed by flow cytometry at the indicated time points. G, MC38 bearing C57BL/6 mice (n = 3 per group) were

p.t. injected with 5 × 105 luciferase expressing iMSCs on day 9. Luciferase signal was analyzed at the indicated time point. H-J, MC38 bearing mice (n = 4 per group) were p.t. injected with

5 × 105 EMSCs on day 9. The frequency of Ki67+GFP+ cells (H), and CD45−GFP+ cells (I) in tumor were analyzed at the indicated time points. J, 200 μg anti-CD8 was administered 1 day before

EMSC injection. Four days after EMSC injection, CD45−GFP+ cells were analyzed in tumor. K, NSG-SGM3 mice were s.c. inoculated with 5 × 105 CT26 tumor cells and 5 × 105 luciferase expressing

iMSCs were p.t. treated on day 10. Five days after MSC treatment, organs were extracted and luciferase signal was analyzed. L, Functional activity of SIL2-EMSC culture supernatant was

assessed by using HEK-BlueTM IL-2 reporter cell assay (n = 2). Data are shown as mean ± s.e.m. from two independent experiments. _P_ value was determined by two-tailed unpaired t test

(B,H,J) or two-way ANOVA (D). Source data are available online. Source data EXTENDED DATA FIG. 4 THERAPEUTIC EFFECT OF IL-2 EXPRESSING ENGINEERED MSC IN VIVO. A, MC38 bearing C57BL/6 mice

(Control, n = 7; EMSC (5x), n = 6; SIL2-EMSC (1x), n = 6; SIL2-EMSC (2x), n = 8; SIL2-EMSC, n = 8) were p.t. treated with 1 × 106 EMSC or SIL2-EMSC. Cells were injected every 3 days from day

9 as many times as indicated. B, MC38 bearing C57BL/6 mice (n = 5 per group) were p.t. treated with different number of EMSC or SIL2-EMSC on day 9 and 12 (black arrow). C, B16 bearing

C57BL/6 mice (n = 5 per group) p.t. treated with 1 × 106 EMSC or SIL2-EMSC on day 7 and 10 (black arrow). CT26 bearing BALB/c mice (n = 5 per group) p.t. treated with 1 × 106 EMSC or

SIL2-EMSC on day 9 and 12 (black arrow). 4T1 bearing BALB/c mice (n = 5 per group) p.t. treated with 1 × 106 EMSC or SIL2-EMSC on day 10 and 13 (black arrow). D,E, Advanced MC38 tumor

bearing C57BL/6J mice (EMSC, n = 5; WTIL2-EMSC, n = 6; SIL2-EMSC, n = 6; EMSC + anti-PD-L1, n = 5; WTIL2-EMSC + anti-PD-L1, n = 7; SIL2-EMSC + anti-PD-L1, n = 6) were treated with 100 μg

α-PD-L1 (i.p.) in combination with 1 × 106 EMSCs, WTIL2-EMSC or SIL2-EMSCs (p.t.) on day 14 and 17 (black arrow). Tumor growth (D) and survival curve (E) are shown. Tumor growth was measured

twice a week. Data are shown as mean ± s.e.m. from two independent experiments. _P_ value was determined by two-way ANOVA (A-D) or log rank test (E). Source data are available online.

Source data EXTENDED DATA FIG. 5 TUMOR-TARGETED PRODUCTION OF SIL2 BY ENGINEERED MSCS PREVENTS SYSTEMIC TOXICITY. A-C, CT26 bearing BALB/c mice (Control, n = 4; SIL2, n = 4; SIL2-EMSC, n = 

5) treated with SIL2-EMSC (1 × 106, p.t.) or SIL2 (20 ug, i.t.) on day 9 and 12. Two days after second treatment, serum, lung, and liver were isolated. A, Livers were extracted and fixed in

10% formalin for 7 days, then H&E staining was performed. Representative example of 10x magnification liver staining from each group is shown. B, Serum level of ALT and AST (B) and

pulmonary wet weight (C) were examined as described in the Methods. D, CT26 bearing BALB/c mice (Control, n = 5; SIL2, n = 6; SIL2-EMSC, n = 6) treated with SIL2-EMSC (1 × 106, p.t.) or SIL2

(20 ug, i.t.) on day 9 and 12 (black arrow). Tumor growth was measured twice a week. E, CT26 bearing BALB/c mice (n = 5 per group) were treated with SIL2-EMSC (1 × 106, p.t.) or SIL2 (20

ug, i.t.) on day 9. IFNγ and TNFα level in the tumor tissue were determined by CBA at 1, 3, or 5 days after treatment. Data are shown as mean ± s.e.m. from two independent experiments. _P_

value was determined by two-tailed unpaired t test (B,C,E) or two-way ANOVA (D). Source data are available online. Source data EXTENDED DATA FIG. 6 EMSC AND SIL2-EMSC DO NOT AFFECT MYELOID

CELL POPULATIONS IN THE TME. A,B, MC38 bearing mice (Control, n = 4; EMSC, n = 4; SIL2-EMSC, n = 6) were p.t. treated with 1 × 106 EMSCs or SIL2-EMSCs on days 9. A, Two days after treatment,

the number of macrophages, dendritic cells, and MDSCs were analyzed. B, Cell surface expression level of CD80 and CD86 on dendritic cells were analyzed. C,D, MC38 bearing mice (n = 5) were

p.t. treated with 1 × 106 EMSCs or SIL2-EMSCs on day 9 and 12. Five days after the second treatment, TILs were analyzed for the number of CD4+ T cells (C) and NK cells (D). E, MC38 bearing

C57BL/6J mice (n = 5) were p.t. 1 × 106 EMSC, SIL2-EMSC or WTIL2-EMSC on day 9 and 12. Five days after the second treatment, TILs were analyzed for the CD8/Treg ratio. Data are shown as mean

± s.e.m. from two or three independent experiments. _P_ value was determined by two-tailed unpaired t test (A-E). Source data are available online. Source data EXTENDED DATA FIG. 7

SIL2-EMSC EXPANDS AND REJUVENATES EXHAUSTED CD8+ TILS. A-C, C57BL/6 J mice (n = 6) bearing MC38 were p.t. treated with 1 × 106 EMSCs or SIL2-EMSCs on day 9 and 12. Five days after the last

treatment, tumor-infiltrating T cells were analyzed for the fold change of number of PD1−TIM3−, PD1+TIM3−, and PD1+TIM3+ CD44+CD8+ T cell subsets (A) and the frequency of CXCR5+TIM3− subset

(B) and CD39+Tcf1− subset (C) in PD1+TOX+CD44+CD8+ T cells in tumor were analyzed. D-F, CT26 bearing BALB/c mice (n = 5) were treated with 1 × 106 EMSC or SIL2-EMSC on day 9 and day 12. Five

days after the second treatment, TILs were analyzed for the number of CD8+ T cells (D) and the frequency of PD1−TIM3−, PD1+TIM3−, and PD1+TIM3+ CD8+ T cell subsets (E) and stem-like

exhausted CD8+ T cells (F). G-I, C57BL/6J mice (n = 5–6) bearing MC38 were p.t. treated with 1 × 106 EMSCs or SIL2-EMSCs on day 9 and 12. Five days after the last treatment, TILs were

analyzed for surface expression level of CD25 and CD122 (G, n = 6 per group), the frequency of Ki67+ cells (H, EMSC, n = 5; SIL2-EMSC, n = 6), and the frequency of active caspase 3+ cells

(I, EMSC, n = 5; SIL2-EMSC, n = 6) in PD1−TIM3−, PD1+TIM3−, and PD1+TIM3+ CD8+ T cell subsets. J, K, MC38 bearing C57BL/6J mice (n = 4) were i.t. treated with single high dose (20 μg) or

prolonged low dose (5 μg, 4 injections) on day 9. Prolonged low dose group were injected with SIL2 every 12 h. CD8+ TILs were analyzed for the frequency of PD1−TIM3−, PD1+TIM3−, PD1+TIM3+

CD8+ T cell subsets and (J) and stem-like CD8+ T cells (K). L, MC38 bearing C57BL/6J mice (n = 4) were treated with SIL2 (10 μg, i.t.) and/or EMSCs (1 × 106, p.t.) and SIL2-EMSC (1 × 106,

p.t.) on day 9 and 12. Five days after treatment, CD8+ TILs were analyzed for the frequency of PD1−TIM3−, PD1+TIM3−, PD1+TIM3+ CD8+ T cell subsets. Data are shown as mean ± s.e.m. from two

or three independent experiments. _P_ value was determined by two-tailed unpaired t test (A–G,I–L). Source data are available online. Source data EXTENDED DATA FIG. 8 SIL2-EMSC EXPANDS AND

FUNCTIONALLY REINVIGORATES EXHAUSTED CD8+ TILS. A, CD8+ TILs from MC38 bearing mice (PD1−TIM3−, n = 3; PD1+TIM3−, n = 6; PD1+TIM3+, n = 6) were co-cultured with irradiated MC38 in the

presence of WTIL2 or SIL2. Two days later, the frequency of CD8+ TIL subsets were analyzed. B,C, CD8+ TILs from MC38 bearing mice (PD1+TIM3−, n = 4; PD1+TIM3+, n = 5) were sorted out and

labeled with CFSE. Cells were co-cultured with irradiated MC38 in the presence or absence of SIL2. Three days later, cell proliferation (B) and apoptosis (C) were determined by flow

cytometry. D, IFN-γ reporter mice (n = 5 per group) bearing MC38 were p.t. treated with 1 × 106 EMSCs or SIL2-EMSCs on day 9. Two days after treatment, tumor-infiltrating T cells were

analyzed to determine the frequency of IFN-γ+ cells in PD1−TIM3−, PD1+TIM3−, and PD1+TIM3+ CD8+ T cell subsets. E, Naive OT-1 splenocytes were activated with OT-1 peptide in the presence of

recombinant IL-2. OT-1 peptide primed OT-1 splenocytes were restimulated with dimerized α-CD3 for 2 d. Representative flow plots of the cell surface phenotype after restimulation (left) and

PD1+TIM3− or PD1+ TIM3+ CD8+ subsets after sorting (right) are shown. F-H, B16-OVA tumor bearing _Rag1_ KO mice were adoptively transferred with OTI CD8+ T cells on day 8. Ten days after

transfer, PD1+TIM3− or PD1+TIM3+ CD8+ TILs were sorted. B16-OVA tumor bearing C57BL/6 mice (EMSC, n = 5; EMSC + PD1+TIM3−, n = 5; EMSC + PD1+TIM3+, n = 5; SIL2-EMSC, n = 6; SIL2-EMSC + 

PD1+TIM3−, n = 5; SIL2-EMSC + PD1+TIM3+, n = 5) were adoptively transferred with 10,000 sorted CD8+ TIL subsets on day 5. EMSC or SIL2-EMSC were treated on day 5 and day 8. F, Experiment

scheme. Tumor growth (G) and survival curve (H) are shown. Data are shown as mean ± s.e.m. from two to three independent experiments. _P_ value was determined by two-tailed unpaired t test

(A-D) or two-way ANOVA (G) or log rank test (H). Source data are available online. Source data EXTENDED DATA FIG. 9 SIL2-EMSCS ARE ELIMINATED BY ADMINISTRATION OF Β-LAP. A, NQO1 expression

in different cancer cell lines and EMSCs was determined by western blotting assay. Representative example from two independent experiments is shown. B, CT26 bearing Balb/c mice were p.t.

injected with 5 × 105 luciferase expressing EMSCs on day 9. Two days after MSC injection, mice were i.t. treated with β-lap (5 mg/kg) every other day for four times. Luciferase signal was

analyzed at indicated time points. C, CT26 bearing BALB/c mice (EMSC, n = 6; EMSC + β-lap, n = 5; SIL2-EMSC, n = 8; SIL2-EMSC + β-lap, n = 6) were i.t. treated with 1 × 106 EMSC or SIL2-EMSC

on day 9 and 12 (black arrow). Two days after MSC injection, mice were i.t. treated with β-lap (15 mg/kg) every other day for four times. Data are shown as mean ± s.e.m. from two

independent experiments. _P_ value was determined by two-way ANOVA (C). Source data are available online. Source data EXTENDED DATA FIG. 10 SIL-EMSC TREATMENT EXHIBITS ANTITUMOR RESPONSES ON

HUMANIZED MOUSE MODEL. A, Cumulative survival in skin cutaneous melanoma (SKCM) patients according to _Tcf7_ expression level in TCGA database. B, TCGA database RNA-seq analysis of the

correlation between the gene expression of _Il2_ and _Tcf7_ in patients with SKCM. C, HCT116 bearing humanized mice (Day 5, n = 5; Day 10, n = 3) were p.t. injected with 5 × 105 EMSCs on day

7. The frequency of CD45−GFP+ cells in tumor was analyzed indicated time points. D, Humanized mice (n = 6 per group) were inoculated with 1 × 106 HCT116 cells and p.t. treated with 1 × 106

EMSCs or SIL2-EMSCs on day 7 and 10 (black arrow). Tumor growth was measured twice a week. E,F, HCT116 bearing humanized mice (EMSC, n = 3; SIL2-EMSC, n = 4) were p.t. treated with 1 × 106

EMSCs or SIL2-EMSCs on day 7 and 10. Five days after the last treatment, tumor-infiltrating T cells were analyzed for number of CD8+ T cells (E), and TCF1+ stem-like CD8+ T cells (F). G,H,

SIL2-EMSCs irradiated with different doses (0–50 Gy). G, Two days after irradiation, cell viability was determined (n = 4). H, Cells were incubated in 1% (hypoxia) O2 for 24 hours. SIL2

levels from cell culture supernatants were determined by ELISA (n = 3). I, C57BL/6 mice (n = 5) were s.c. inoculated with 1 × 106 MC38 cells and p.t. treated with 1 × 106 EMSCs, SIL2-EMSCs,

or SIL2-EMSC irradiated with 20 Gy on days 9 and 12 (indicated by black arrow). Tumor growth was measured twice a week. Data are shown as mean ± s.e.m. from two independent experiments.

Statistical analysis of TCGA data were performed by log-rank test (A) or Spearman’s rho correlation test (B). _P_ value was determined by two-way ANOVA (D,I) or two-tailed unpaired t test

(E, F). Source data are available online. Source data SUPPLEMENTARY INFORMATION SUPPLEMENTARY INFORMATION Supplementary Fig. 1. REPORTING SUMMARY SUPPLEMENTARY TABLE Supplementary Table 1.

List of differentially expressed genes from scRNA-seq. Supplementary Table 2. List of antibodies used in this study. SOURCE DATA SOURCE DATA FIG. 1 Statistical source data. SOURCE DATA FIG.

2 Statistical source data. SOURCE DATA FIG. 3 Statistical source data. SOURCE DATA FIG. 4 Statistical source data. SOURCE DATA FIG. 5 Statistical source data. SOURCE DATA FIG. 6 Statistical

source data. SOURCE DATA FIG. 7 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 1 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 2 Statistical source data. SOURCE DATA

EXTENDED DATA FIG. 3 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 4 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 5 Statistical source data. SOURCE DATA EXTENDED DATA

FIG. 6 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 7 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 8 Statistical source data. SOURCE DATA EXTENDED DATA FIG. 9

Statistical source data. SOURCE DATA EXTENDED DATA FIG. 9 Unprocessed western blots. SOURCE DATA EXTENDED DATA FIG. 10 Statistical source data. RIGHTS AND PERMISSIONS Springer Nature or its

licensor (e.g. a society or other partner) holds exclusive rights to this article under a publishing agreement with the author(s) or other rightsholder(s); author self-archiving of the

accepted manuscript version of this article is solely governed by the terms of such publishing agreement and applicable law. Reprints and permissions ABOUT THIS ARTICLE CITE THIS ARTICLE

Bae, J., Liu, L., Moore, C. _et al._ IL-2 delivery by engineered mesenchymal stem cells re-invigorates CD8+ T cells to overcome immunotherapy resistance in cancer. _Nat Cell Biol_ 24,

1754–1765 (2022). https://doi.org/10.1038/s41556-022-01024-5 Download citation * Received: 04 November 2021 * Accepted: 27 September 2022 * Published: 06 December 2022 * Issue Date: December

2022 * DOI: https://doi.org/10.1038/s41556-022-01024-5 SHARE THIS ARTICLE Anyone you share the following link with will be able to read this content: Get shareable link Sorry, a shareable

link is not currently available for this article. Copy to clipboard Provided by the Springer Nature SharedIt content-sharing initiative