AUTHORS
Hao Yan, PhD, Xiangyue Zhang, PhD2, Jeanette Baker3, Guotao Yu, PhD4, Yanlan Wang, PhD5, Edgar G Engleman, MD, PhD6, Robert Negrin, MD7
AFFILIATIONS
1 Stanford University School of Medicine, Palo Alto, California, United States
2 Stanford University School of Medicine, Palo Alto, California, United States
3 Stanford University School of Medicine, Palo Alto, California, United States
4 Stanford University School of Medicine, Palo Alto, California, United States
5 Stanford University School of Medicine, Palo Alto, California, United States
6 Stanford University School of Medicine, Palo Alto, California, United States, Stanford University, Palo Alto, California, United States
7 Stanford University School of Medicine, Palo Alto, California, United States
ABSTRACT
Graft-versus-host disease (GVHD) remains a major cause of morbidity and mortality after allogeneic hematopoietic cell transplantation (HCT). The administration of apoptotic cells generated by extracorporeal photopheresis (ECP) can ameliorate GVHD in patients and mouse models. Although dendritic cells (DCs) are known to contribute to this response, the specific DC subset and molecular mechanism underlying the protective effects of ECP remain unclear. cDC1s are specialized antigen-presenting cells that acquire cell-associated antigens by engulfing apoptotic cells (ACs) through efferocytosis or by taking up necrotic cells, thereby regulating antigen-specific T cell tolerance and immunity, respectively. The Engleman laboratory recently showed that erythropoietin receptor (EPOR) on cDC1s dictates immune tolerance, and that the tolerogenic maturation of cDC1s following efferocytosis is EPOR-dependent and modulated by the level of EPO-EPOR signaling. We therefore hypothesized that co-administration of EPO with ECP-derived apoptotic cells would enhance protection against GVHD and that this effect would require EPOR signaling in cDC1s.
Splenocytes were treated ex vivo with 8-MOP and UVA irradiation to generate ECP-treated ACs. Recipients received 1×107 ACs on day -2 before HCT. Recombinant hEPO was administered intraperitoneally at 50 IU/mouse/day from day -3 to day +1. Mice were conditioned with total body irradiation (TBI: BALB/c: 880 cGy, C57BL/6: 1100 cGy) and received T-cell-depleted bone marrow (TCD-BM; 5x106/animal) plus conventional T cells (Tcon; BALB/c, 1x106/animal; C57BL/6, 4x106/ animal) from allogeneic donors. Batf3-/- mice were used to assess the role of cDC1. EPOR-tdTomato reporter mice were used to evaluate EPOR expression and intensity. To define the role of EPOR signaling in cDC1, cDC1-specific EPOR conditional knockout mice (EporΔXCR1; Xcr1Cre/+Eporflox/flox) were generated to determine the requirement for cDC1-intrinsic EPOR signaling.
To investigate the role of cDC1 in GVHD, we used Batf3−/− C57BL/6 recipients in aGVHD model. Batf3−/− mice, that lack cDC1, developed more severe GVHD than WT, with reduced survival and increased clinical scores. Moreover, the GVHD protective effect of ECP was markedly attenuated in Batf3−/− recipients, suggesting that cDC1 are required for ECP-mediated allo-T cell suppression. Using Epor-tdTomato reporter mice, we found that ECP increased the frequency of cDC1s and their EPOR expression, effects further enhanced by EPO co-administration. Serum EPO levels increased following TBI and were further elevated by ECP, suggesting activation of an endogenous EPO-EPOR tolerogenic axis in cDC1s. In C57BL/6 to BALB/c aGVHD mouse model, animals receiving Tcon alone succumbed to GVHD, whereas ECP prolonged survival. EPO alone showed little protective effect, indicating that active efferocytosis is necessary for GVHD prevention. Notably, ECP+EPO resulted in much improved survival compared with either treatment alone, with median survival not reached (p<0.0001). Importantly, this combination did not diminish graft versus tumor activity. In EporΔXcr1 recipients, the protective benefits of ECP and ECP+EPO were both diminished, demonstrating that EPOR signaling in cDC1s is required for ECP or ECP+EPO-mediated GVHD prevention. To assess whether ECP induces EPOR upregulation on cDC1s in patients, we performed a preliminary translational study in which 10 mL peripheral blood samples were collected from GVHD patients before and 3–4 days after ECP treatment. These preliminary results indicate that ECP treatment is associated with increased EPOR expression on circulating CD141+ cDC1s in patients with a corresponding expansion of FOXP3+ Tregs.
Together, our results demonstrate that EPO-EPOR signaling in cDC1s is critical for ECP-induced immune tolerance in the setting of allogeneic HCT. Short-term EPO administration enhances ECP-mediated GVHD prevention through the EPO/EPOR-cDC1 axis, suggesting ECP+EPO as a novel strategy to suppress GVHD.
DISCLOSURES
Hao Yan, PhD: No relevant conflicts of interest to declare.