The phagocytic clearance of the large fraction of transplanted cells that typically undergo rapid cell death contributes to therapeutic outcomes.
Cell therapies have produced mixed clinical outcomes. Haematopoietic stem cells, chimeric antigen receptor T cells and other cell products with a well-defined mechanism of action (MOA) tend to perform well; but those cell products with poorly defined MOAs have had limited success. In particular, mesenchymal stromal cells (MSCs) have been a subject of intense focus over the past two decades, with over 700 clinical trials reported across a range of diseases. Overall, MSCs have a positive safety profile, yet the majority of efficacy studies of MSC transplantation therapies have failed to deliver1,2,3,4. Even for the few MSC products that have been clinically approved (for treating graft-versus-host disease (GvHD), complex perianal fistulas and spinal-cord injury, in particular), their overall clinical efficacy and adoption remain unclear2,5,6. Also, their unapproved use in clandestine stem cell clinics has led to hype and to provocative debate within scientific communities and the public regarding whether these therapies work and whether they should be excluded from scientific forums, peer-reviewed journals and research funding7,8,9.
The mixed clinical outcomes and misconceptions surrounding cell therapies can be largely attributed to an inadequate understanding of the fate of transplanted cells and of how they exert therapeutic outcomes. Historically, the multipotency of MSCs (and their branding as ‘stem cells’10) drove the belief that these cells assisted regeneration in the host through the differentiation and direct replacement of injured cells (Fig. 1a). This interpretation has been largely dismissed, because although MSCs can be made to differentiate through artificial cues in vitro, there is scant evidence that this can occur in vivo11. The concept that functional outcomes from transplanted MSCs are mediated by the cells’ secretome was later popularized, partly because transplanted MSCs sytemically localize to sites of inflammation or injury, and because they establish regenerative and immunosuppressive microenvironments through cell–cell crosstalk and through the release of paracrine factors (such as indoleamine 2,3-dioxygenase; IDO)11,12,13 (Fig. 1b). However, the potential paracrine mechanisms of MSCs have been questioned by the observation that most MSCs transplanted systemically (the administration route most commonly used in clinical trials) are unable to progress past the microvasculature of filter organs (such as the lungs), owing to their large size; in fact, less than 1% of the transplanted cells reach and engraft at target sites14,15,16,17,18,19,20,21. Indeed, on intravenous infusion most MSCs die and are cleared from the body within 24 hours14,15,16,17,18. It is also worth noting that MSC-homing studies can be confounded by the observation that imaging agents, such as fluorescent dyes labelling the transplanted cells, can be transferred to endogenous phagocytes in vivo14, and therefore cannot faithfully determine the biodistributions or pharmacokinetics of live MSCs. Nevertheless, despite their short lifespan in vivo, MSCs are capable of conferring long-lasting therapeutic outcomes across various preclinical models22. This suggests that there are alternative mechanisms underlying their therapeutic function.
It has been suggested that the clearance of dying and dead cells through phagocytosis by the host reticuloendothelial system2,18,23,24,25,26,27,28,29,30,31,32,33,34 (Fig. 1c) may be of functional relevance. The phagocytic clearance of endogenous dying cells and their cellular debris (so-called efferocytosis) is an integral component of tissue homeostasis and repair35,36,37,38. It involves the recruitment of a subset of phagocytes to sites of cell death through chemical signals (such as nucleotides released by dying cells), the recognition of damaged cells through ‘eat me’ signals (such as exposed inner-membrane lipid phosphatidylserine), and the engulfing of the cells35,36,37,38. The phagocytic clearance of dead cells has a profound impact on the responses of both the innate and adaptive immune system in the context of resolving inflammation (which leads to tolerance induction through phenotype changes in immune cells) and of restoring the function of injured tissues. These dying-cell-clearance pathways are currently being explored as therapeutic targets for treating inflammation and diseases associated with it35,36,37,38.
Transplanted cells rapidly undergo apoptosis, necrosis or other death mechanisms39 as a result of physical factors (for example, mechanical lysis in the microvascular bed of the lung) and of active biochemical factors (such as apoptosis induced by host cytotoxic cells, by the complement system and by autophagy). By releasing extracellular vesicles (EVs), organelles (such as mitochondria40), bioparticles, apoptotic bodies, heat-shock proteins and other damage-associated molecular-pattern constituents (such as high-mobility group box 1 (HMGB1), mitochondrial DNA, and ATP), dying cells mimic a ‘local sterile injury’ or inflammatory environment41 that can act on tissue-resident phagocytes, recruited phagocytes and circulating phagocytes such as macrophages, monocytes, neutrophils and dendritic cells. Whereas systemically transplanted cells largely disintegrate in the lungs, their cellular remnants can be transported via the bloodstream to other parts of the body, where they are phagocytosed (for example by Kupffer cells in the liver32).
The clearance of transplanted cells polarizes phagocytes toward anti-inflammatory, tolerogenic and regenerative phenotypes, which ultimately contribute to the observed therapeutic effects. Indeed, these dead-cell-clearing reprogrammed phagocytes can migrate throughout the body to further mediate tissue repair and regeneration at sites of disease directly, or by modulating the cells of the adaptive immune system through changes in immunosuppression or in the antigen-presentation machinery32. For instance, intravenously transfused mouse bone-marrow-derived MSCs can attenuate sepsis in mice by polarizing endogenous macrophages to produce the anti-inflammatory cytokine interleukin-10 (IL-10)42. Although this study42 associated the therapeutic effect of transplanted MSCs to changes in endogenous innate immune-cell populations, the proposed mechanism involved the active crosstalk between MSCs and macrophages via prostaglandin E2 (PGE2) signalling, and did not explicitly implicate the clearance of dying cells, which could also produce PGE2 and mediate tissue repair43. Intriguingly, a more recent study showed that heat-inactivated ‘dead’ MSCs (incapable of metabolic, proliferative or secretory activity) maintained their immunomodulatory capacity in a mouse sepsis model33. This indicates that the therapeutic effect of MSCs in this context was independent of cellular activity, and that it probably involved the passive recognition and phagocytosis of the MSCs by monocytic cells.
However, human umbilical-cord-derived MSCs can skew human CD14+/CD16– classical monocytes toward an immunosuppressive and tolerant CD14+CD16+CD206+ phenotype with increased expression of programmed-death ligand-1 (PD-1) and IL-10 as well as with reduced expression of tumour necrosis factor α (ref. 32). These ‘reprogrammed’ monocytes induced the differentiation of regulatory T cells in mixed lymphocyte reactions. The majority of tail-vein-infused MSCs died, and were rapidly cleared within 24 hours by innate immune cells (which acquired regulatory phenotypes following the phagocytosis of the MSCs) within the lungs, liver and blood. This study also confirmed that the phagocytic clearance of MSCs can induce functional changes in monocytes, yet implied that the redistribution of the monocytes to various body sites could enable these cells to further interface with the adaptive immune system and modulate it32. Nonetheless, whether the phagocytosis of MSCs could improve clinical outcomes in models of injury or disease remained unclear. Yet in a GvHD model (created via the induction of IDO and PGE2 production in endogenous phagocytes), transplanted apoptotic MSCs generated ex vivo were immunosuppressive34,44. Also, the transplanted MSCs were actively induced to undergo perforin-dependent apoptosis by recipient cytotoxic CD8+ T cells and CD56+ natural killer cells, and recipient phagocytes that engulfed these apoptotic MSCs produced increased levels of IDO, which was essential to the immunosuppression observed in the GvHD model. Moreover, patients with GvHD who exhibited high cytotoxic activity against MSCs responded well to MSC infusion, whereas those with low activity benefited less34. Similar macrophage phagocytosis-mediated mechanisms were also found to be responsible for the therapeutic effects of MSCs in animal models of asthma25,26 and of cardiovascular diseases28,29,30. In aggregate, these studies support the idea of an autonomous cellular-effector mechanism involving phagocytosis and the clearance of dead cells in addition to multilineage differentiation and the paracrine signalling of live transplanted cells. This idea is further supported by the observation that MSC-derived EVs, including exosomes and membrane-derived nanoparticles that carry biologically active RNAs and proteins, can exert therapeutic functions in preclinical models by targeting recipient phagocytic cells45,46,47.
Implications for cell therapies
The biological processes responsible for clearing dying cells may shed some light on open questions arising from discrepancies between results from in vitro studies and preclinical work, and observations from human studies. The mechanisms of phagocytic clearing could explain the prolonged efficacy of otherwise short-lived MSCs despite lack of engraftment at diseased sites (as observed across many different disease settings, irrespective of the MSC source and preparation16,48). Indeed, MSCs derived from bone marrow, adipose tissue or the umbilical cord as xenogeneic, allogeneic or syngeneic forms have all shown some degree of efficacy in a large variety of preclinical disease models (for instance, models of GvHD, cartilage defects, ischaemic stroke, multiple sclerosis, Crohn’s disease and myocardial infarction). Conversely, outcomes in a clinical setting have been poor, and can be attributed to lack of control, predictability and robustness, all potentially associated with a ‘non-specific’ dying-cell-clearance mechanism. This hypothesis would help resolve the hotly debated7,8,9 ‘stem cell myth’, and support the development of a holistic view of a MOA for these cell therapies.
We recommend a careful re-examination of observations in cell-therapy studies in terms of their preclinical and clinical design, to assess the potential involvement of the dying-cell clearance mechanism, and, if needed, the implementation of revised MOAs, potency assays and patient-stratification strategies for improving clinical outcomes. Furthermore, the universal and non-specific nature of this mechanism might be exploited for using MSCs as prophylactic immunomodulatory interventions for preventing inflammatory conditions. We caution that a better understanding of efferocytosis mechanisms is critical to avoid or minimize any detrimental effects from cell therapies; in particular, under defective apoptotic cell clearance the therapies may be responsible for transient inflammatory responses49 or for long-lasting autoimmune and chronic inflammatory disorders35,36,37,38.
Although our discussion of these considerations has focused on intravenously delivered MSCs, the dying-cell clearance mechanism can be broadly applicable to other cell therapies involving alternative cell sources, disease models or administration routes (except when their MOAs are well defined, as in the case of haematopoietic stem cells and of chimeric antigen receptor T cells). For instance, the inflammatory condition instant blood-mediated inflammatory reaction, which has been observed following the transplantation of islets of Langerhans, hepatocytes and MSCs50 and can lead to cell destruction, can potentially mediate beneficial effects in cell therapies18,51. Similarly, the vast majority of exogenously transplanted neural stem cells (NSCs) for the treatment of disorders of the central nervous system die within a short period of time regardless of their administration route (systemic infusion or local injection), and there is limited evidence of engraftment and cell replacement52,53,54. It would, thus, be worth examining how phagocytes such as microglia in the brain are involved in the clearance of dead NSCs and whether or not that accounts for the observed beneficial therapeutic effects.
A cell-clearance pathway for therapy
Owing to their plasticity and central role in immunomodulation and tissue regeneration55,56,57, macrophages and antigen-presenting cells can be modulated by harnessing cell efferocytosis58,59 to maximize a desirable therapeutic outcome. Transplanted cells can be engineered with pro-inflammatory stimuli60, tolerogenic antigens or immunomodulatory agents61 so that a subset of phagocytes can potentially be reprogrammed on engulfing the modified dying cells. In this way, phagocytes may be modulated to become pro-inflammatory, tolerogenic or regenerative, for treating cancers62, for attenuating autoimmune disorders61 and transplant rejection58,63, or to promote wound healing58,63, respectively (Fig. 2a). For instance, glucocorticoid-containing nanoparticles could be loaded into exogenous cells to facilitate in vivo phagocytosis by both inducing apoptosis and enhancing engulfment38,64. Through phagocytosis, macrophages can also be modified with the knockdown of the signal regulatory protein α, to express tumour-killing cytokines (such as interferon α) or to activate M1-polarizing transcription factors (as shown preclinically for cancer therapies65).
Synthetic EVs66,67,68,69, nanocarriers62,70,71 and apoptotic-cell mimics34,35,36,37,38,72,73 could also be exploited as cell-free therapeutics for the modulation of innate phagocytes (Fig. 2b). For instance, we recently showed that MSC-derived exosomes had therapeutic effects in a mouse model of multiple sclerosis (experimental autoimmune encephalomyelitis) by suppressing neuroinflammation and upregulating regulatory T cells45. However, the therapeutic effects of cell-derived vesicles often require the involvement of ‘intermediate’ phagocytic cell types46,74,75,76,77,78,79. And, as with cell therapies, the effectiveness of EVs and their MOAs depends on EV origin, preparation and route of administration66,68. For example, phosphatidylserine-presenting liposomes that mimicked anti-inflammatory apoptotic cells modulated cardiac macrophages to mediate infarct repair70. Moreover, nanocarriers co-delivering immunomodulatory and tolerogenic agents (such as transforming growth factor beta (TGF-β) or rapamycin) and disease-specific antigens (such as insulin peptides in the context of type 1 diabetes) to dendritic cells induced immune tolerance for the treatment of autoimmune disorders61. And multicellular therapy comprising macrophages and MSCs co-cultured ex vivo has been evaluated in a clinical setting80, as combination cell therapies may support a wider range of biological activities (potentially also efferocytosis).
Paracrine signals versus efferocytosis
Employing cell death and clearance mechanisms could thus complement the conventional MOAs of cell therapies, but the effects of efferocytosis may be variable and context-dependent. For example, apoptotic MSCs ameliorated GvHD when infused intraperitoneally, but exhibited greatly reduced efficacy when administered intravenously34. Other work has shown that the viability of MSCs is required for their therapeutic outcome in a mouse model of colitis induced by dextran sulfate sodium, in which heat-killed MSCs or MSCs administered immediately post-thaw were shown to be ineffective81. The MOA of MSCs in this colitis model seems to be based on polarizing IL-10-producing tissue macrophages, probably via paracrine signalling, which leads to systemic effects despite the compartmentalized delivery of the MSCs82. In addition, numerous in vitro experiments with purified effector cells and without monocytes showed immunomodulatory effects of MSCs on both innate and adaptive cell systems without implicating efferocytosis83. Furthermore, a recent phase-2a clinical trial of the use of allogeneic MSCs for the treatment of acute respiratory distress syndrome implied that the effectiveness of intravenously transfused MSCs is likely to depend on cell viability and fitness84. These findings speak to the essential role of paracrine functionality in certain settings; hence, paracrine signalling should remain a key component of the MOA of MSCs.
Table 1 lists studies that specifically assessed living cells versus dead cells in their therapeutic effect in similar settings and that implicated whether paracrine or dead-cell-clearance mechanisms are favoured. The apparent discrepancies in the literature clearly highlight the complexity, variability and the context-dependent nature of each MOA. In particular, the route of administration and the in vivo persistence (engraftment) of transplanted cells have recently been identified as key determinants to their MOA85,86. Whereas efferocytic clearance is mostly relevant following intravenous transfusions, when MSCs die and are cleared in filter organs, paracrine mechanisms are likely to be more important in the context of the extravascular delivery of live MSCs with sustained persistence and engraftment87. In agreement with this rationale, systemically administered MSCs can enhance wound repair distally from the lungs, where the cells become entrapped, whereas locally administered MSCs can redistribute within the wound bed and lead to wound repair86. Moreover, the therapeutic potency of MSCs in the induced colitis model is also dependent on the administration route81: subcutaneous and intraperitoneal MSC administration, which can bypass the pulmonary filter, are more effective than the maximally tolerated intravenous dose for reducing colitis severity. In this study81, the therapeutic outcomes of the MSCs were associated with sustained in vivo persistence (up to ten days) in the cases of subcutaneous or intraperitoneal administration of fresh live MSCs, whereas in vivo engraftment was severely compromised for dead MSCs and for post-thawed MSCs. Also, clinically approved human adipose-derived MSCs (Alofisel) indicated for the treatment of Crohn’s fistular disease are delivered subdermally6; therapeutic outcomes involving the in vivo persistence of live MSCs would support the paracrine MOA in these settings.
Multiple mechanisms are likely to be at play for cell therapies, and it will be important in future work to quantitatively determine the relative contribution of each mechanism in a given disease setting. An indirect way to measure the contribution of efferocytosis in published studies (Table 1) is to compare the therapeutic effects of live cells to their apoptotic counterpart prepared in vitro (still, such ‘dead’ or non-viable MSCs may not fully recapitulate the active, host-driven efferocytosis process that takes place in vivo). It is also important to determine the commonality and variability of the efferocytosis mechanism among and within different settings: it is expected that efferocytosis-mediated therapeutic efficiency and duration may be dependent on the characteristics of various therapeutic cells (source, preparation, size and dose), administration routes (which determine the sites of action), disease setting and immune status of the host (such as the presence and characteristics of different subtypes of phagocytes). In this context, low-passage MSCs, which are smaller in size and therefore exhibit reduced lung entrapment, elicited minimal innate immune responses and exhibited improved survival and function in vivo when compared with their high-passage, culture-expanded counterparts49,51,88. Therefore, to enhance a paracrine-mediated mechanism (rather than efferocytosis), cell preparation and engineering strategies that improve in vivo survival, homing89,90 and engraftment should be considered. ‘Engineered survival’ can indeed lead to enhanced therapeutic efficacy. Moreover, although the intravenous route is the most commonly used means of administration of MSCs in clinical trials, recent evidence6,81,82,85,87 suggests that extravascular delivery can lead to sustained persistence and to clinical effects, which should be further examined. It would be particularly interesting to examine whether, and to what extent, efferocytosis exerts therapeutic function in the case of compartmentalized cell administration; after all, locally injected cells die and can also be cleared by phagocytes34. Hence, mechanistic studies should be conducted to validate cell death and efferocytosis as a critical MOA in cell therapies, to facilitate the development of potent in vitro assays, patient stratification, and, ultimately, robust clinical translation.
Martin, I., Galipeau, J., Kessler, C., Blanc, K. L. & Dazzi, F. Sci. Transl. Med. 11, eaat2189 (2019).
Galipeau, J. & Sensébé, L. Cell Stem Cell 22, 824–833 (2018).
Bianco, P. et al. Nat. Med. 19, 35–42 (2013).
Fibbe, W. E., Dazzi, F. & LeBlanc, K. Nat. Med. 19, 812–813 (2013).
Cyranoski, D. Nature 565, 544–545 (2019).
Panés, J. et al. Lancet 388, 1281–1290 (2016).
Sipp, D., Robey, P. G. & Turner, L. Nature 561, 455–457 (2018).
Caplan, A. I. Nature 566, 39 (2019).
Galipeau, J., Weiss, D. J. & Dominici, M. Cytotherapy 21, 1–2 (2019).
Caplan, A. I. J. Orthop. Res. 9, 641–650 (1991).
Caplan, A. I. Stem Cells Transl. Med. 6, 1445–1451 (2017).
Singer, N. G. & Caplan, A. I. Annu. Rev. Pathol. 6, 457–478 (2011).
Uccelli, A., Moretta, L. & Pistoia, V. Nat. Rev. Immunol. 8, 726–736 (2008).
Leibacher, J. & Henschler, R. Stem Cell Res. Ther. 7, 7 (2016).
Parekkadan, B. & Milwid, J. M. Annu. Rev. Biomed. Eng. 12, 87–117 (2010).
Ankrum, J. & Karp, J. M. Trends Mol. Med. 16, 203–209 (2010).
Salvadori, M. et al. Mol. Ther. Methods Clin. Dev. 14, 1–15 (2019).
Moll, G. et al. Adv. Exp. Med. Biol. 951, 77–98 (2016).
Fischer, U. M. et al. Stem Cells Dev. 18, 683–691 (2009).
Eggenhofer, E. et al. Front. Immunol. 3, 297 (2012).
von Bahr, L. et al. Stem Cells 30, 1575–1578 (2012).
Liu, L. et al. EBioMedicine 47, 563–577 (2019).
Ghahremani Piraghaj, M. et al. Life Sci. 212, 203–212 (2018).
Eggenhofer, E., Luk, F., Dahlke, M. H. & Hoogduijn, M. J. Front. Immunol. 5, 148 (2014).
Braza, F. et al. Stem Cells 34, 1836–1845 (2016).
Mathias, L. J. et al. J. Immunol. 191, 5914–5924 (2013).
Jiang, Y. et al. Basic Res. Cardiol. 105, 431–442 (2010).
Lu, W. et al. Int. J. Cardiol. 165, 333–340 (2013).
Ben-Mordechai, T. et al. J. Am. Coll. Cardiol. 62, 1890–1901 (2013).
Thum, T., Bauersachs, J., Poole-Wilson, P. A., Volk, H.-D. & Anker, S. D. J. Am. Coll. Cardiol. 46, 1799–1802 (2005).
Carty, F., Mahon, B. P. & English, K. Clin. Exp. Immunol. 188, 1–11 (2017).
de Witte, S. F. H. et al. Stem Cells 36, 602–615 (2018).
Luk, F. et al. Stem Cells Dev. 25, 1342–1354 (2016).
Galleu, A. et al. Sci. Transl. Med. 9, eaam7828 (2017).
Abdolmaleki, F. et al. Front. Immunol. 9, 1645 (2018).
Nagata, S., Hanayama, R. & Kawane, K. Cell 140, 619–630 (2010).
Boada-Romero, E., Martinez, J., Heckmann, B. L. & Green, D. R. Nat. Rev. Mol. Cell Biol. 21, 398–414 (2020).
Poon, I. K. H., Lucas, C. D., Rossi, A. G. & Ravichandran, K. S. Nat. Rev. Immunol. 14, 166–180 (2014).
Weiss, D. J. et al. Front. Immunol. 10, 1191 (2019).
Jackson, M. V. et al. Stem Cells 34, 2210–2223 (2016).
Fleshner, M. & Crane, C. R. Trends Immunol. 38, 768–776 (2017).
Németh, K. et al. Nat. Med. 15, 42–49 (2009).
Jäger, R. & Fearnhead, H. O. Biochem. Res. Int. 2012, e453838 (2012).
Cheung, T. S., Galleu, A., von Bonin, M., Bornhäuser, M. & Dazzi, F. Haematologica 104, e438–e441 (2019).
Riazifar, M. et al. ACS Nano 13, 6670–6688 (2019).
Gonçalves, Fd. C. et al. Sci. Rep. 7, 12100 (2017).
Phinney, D. G. et al. Nat. Commun. 6, 8472 (2015).
Lalu, M. M. et al. PLoS ONE 7, e47559 (2012).
Moll, G. et al. Stem Cells 30, 1565–1574 (2012).
Moll, G. et al. Trends Mol. Med. 25, 149–163 (2019).
Moll, G. & Blanc, K. L. ISBT Sci. Ser. 10, 357–365 (2015).
Hicks, A. U., MacLellan, C. L., Chernenko, G. A. & Corbett, D. Brain Res. 1231, 103–112 (2008).
Liao, L.-Y., Lau, B. W.-M., Sánchez-Vidaña, D. I. & Gao, Q. Neural Regen. Res. 14, 1129–1137 (2019).
Donegà, M., Giusto, E., Cossetti, C. & Pluchino, S. in Neural Stem Cells: New Perspectives https://doi.org/10.5772/55426 (IntechOpen, 2013).
Vannella, K. M. & Wynn, T. A. Annu. Rev. Physiol. 79, 593–617 (2017).
Wynn, T. A. & Vannella, K. M. Immunity 44, 450–462 (2016).
Mosser, D. M. & Edwards, J. P. Nat. Rev. Immunol. 8, 958–969 (2008).
Korns, D. R. P., Frasch, S. C. P., Fernandez-Boyanapalli, R. P., Henson, P. M. P. & Bratton, D. L. M. Front. Immunol. 2, 57 (2011).
Spiller, K. L. & Koh, T. J. Adv. Drug Deliv. Rev. 122, 74–83 (2017).
Yin, J. Q., Zhu, J. & Ankrum, J. A. Nat. Biomed. Eng. 3, 90–104 (2019).
Stabler, C. L., Li, Y., Stewart, J. M. & Keselowsky, B. G. Nat. Rev. Mater. 4, 429–450 (2019).
Tuettenberg, A., Steinbrink, K. & Schuppan, D. Nanomedicine 11, 2735–2751 (2016).
Smith, T. D., Nagalla, R. R., Chen, E. Y. & Liu, W. F. Adv. Drug Deliv. Rev. 114, 193–205 (2017).
Ankrum, J. A., Dastidar, R. G., Ong, J. F., Levy, O. & Karp, J. M. Sci. Rep. 4, 4645 (2014).
Zhang, F. et al. Nat. Commun. 10, 3974 (2019).
Phinney, D. G. & Pittenger, M. F. Stem Cells 35, 851–858 (2017).
Théry, C., Ostrowski, M. & Segura, E. Nat. Rev. Immunol. 9, 581–593 (2009).
Riazifar, M., Pone, E. J., Lötvall, J. & Zhao, W. Annu. Rev. Pharmacol. Toxicol. 57, 125–154 (2017).
Armstrong, J. P. K., Holme, M. N. & Stevens, M. M. ACS Nano 11, 69–83 (2017).
Harel-Adar, T. et al. Proc. Natl Acad. Sci. USA 108, 1827–1832 (2011).
Bagalkot, V., Deiuliis, J. A., Rajagopalan, S. & Maiseyeu, A. Adv. Drug Deliv. Rev. 99, 2–11 (2016).
Saas, P., Daguindau, E. & Perruche, S. Stem Cells 34, 1464–1473 (2016).
Pilon, C. et al. Front. Immunol. 10, 2908 (2019).
Zhang, B. et al. Cytotherapy 20, 687–696 (2018).
Shigemoto-Kuroda, T. et al. Stem Cell Rep. 8, 1214–1225 (2017).
Lankford, K. L. et al. PLoS ONE 13, e0190358 (2018).
Henao Agudelo, J. S. et al. Front. Immunol. 8, 881 (2017).
Song, Y. et al. Stem Cells 35, 1208–1221 (2017).
Domenis, R. et al. Sci. Rep. 8, 13325 (2018).
Patel, A. N. et al. Lancet 387, 2412–2421 (2016).
Giri, J. & Galipeau, J. Blood Adv. 4, 1987–1997 (2020).
Giri, J., Das, R., Nylen, E., Chinnadurai, R. & Galipeau, J. Cell Rep. 30, 1923–1934.e4 (2020).
Shi, Y. et al. Nat. Rev. Nephrol. 14, 493–507 (2018).
Matthay, M. A. et al. Lancet Respir. Med. 7, 154–162 (2019).
Caplan, H. et al. Front. Immunol. 10, 1645 (2019).
Kallmeyer, K. et al. Stem Cells Transl. Med. 9, 131–144 (2019).
Braid, L. R., Wood, C. A., Wiese, D. M. & Ford, B. N. Cytotherapy 20, 232–244 (2018).
Le Blanc, K. et al. Lancet 363, 1439–1441 (2004).
Segaliny, A. I. et al. EBioMedicine 45, 39–57 (2019).
Liao, W. et al. Biomaterials 77, 87–97 (2016).
Gupta, N. et al. J. Immunol. 179, 1855–1863 (2007).
Kavanagh, H. & Mahon, B. P. Allergy 66, 523–531 (2011).
Sun, J. et al. Cell. Physiol. Biochem. 27, 587–596 (2011).
Chang, C.-L. et al. J. Transl. Med. 10, 244 (2012).
We thank J. Gu for helpful comments. This work was supported by the NIH grants 1DP2CA195763, R21CA219225 and P2CHD086843 (the Alliance for Regenerative Rehabilitation Research & Training; AR3T), and by DOD/CDMRP W81XWH-17-1-0522 and NSF/SBIR no. 1913404.
W.Z. is a co-founder of Velox Biosystems Inc., Amberstone Biosciences Inc. and Arvetas Biosciences Inc.
About this article
Cite this article
Wagoner, Z.W., Zhao, W. Therapeutic implications of transplanted-cell death. Nat Biomed Eng 5, 379–384 (2021). https://doi.org/10.1038/s41551-021-00729-6