Exosomes, PD-L1 and aGvHD: Perspectives for WJMSC-mediated Therapy

Tumor-derived small extracellular vesicles or exosomes which carry the checkpoint PD-L1 are directly involved in immune evasion and uncontrolled tumor growth. We have recently reported that PD-L1 is also enriched on Wharton’s Jelly Mesenchymal Stromal Cell (WJMSC)-associated exosomes [1]. This biological property is likely associated with the physiologic role of WJMSCs at the maternal-fetal interface. These cells have been shown to protect the developing fetus from attack by the maternal immune system due to the foreign paternal antigens which the fetus express. In this manuscript, we review the important roles of exosome-enriched PD-L1 (exo-PDL1) during WJMSCmediated therapy for acute graft-versus-host disease (aGvHD). We focus on exo-PD-L1’s immunomodulation on T cell receptor (TCR) signaling in CD4+ T cells. In addition, we extend our discussions to other exosomeassociated immune regulators, highlighting the potential to develop a cell-free, WJMSC-derived exosome therapy to treat complex immune diseases.


Exo-PD-L1 Underlies WJMSC-mediated aGvHD Therapy
Allogeneic hematopoietic stem cell transplantation (HSCT) provides a reliable treatment for various hematological malignancies [2]. Although the donor's immune cells help eliminate the tumor cells (an effect of graft-versus leukemia, GVL), they also cause the serious complication of acute graft-versus-host disease (aGvHD) [3,4]. aGvHD occurs due to donor T cells recognizing and attacking the recipient's non-malignant tissues, impacting the clinic application of HSCT for patients [2,5].
Immunomodulatory mesenchymal stem cells (MSCs) have been widely used to treat aGvHD and other immune diseases [6,7]. We recently reported the first investigatorinitiated clinical trial investigating the safety and efficacy of WJMSC in steroid-refractory GvHD [8]. To date, there were at least 25 registered clinical trials for the exploration of the treatment of MSCs for aGvHD patients and 122 for other immunological diseases (http://clinicaltrials.gov/; accessed February 2021). One of the important molecular mechanisms lies in the immune regulatory checkpoint PD-L1. It is well known that checkpoint PD-L1 signal suppresses the T cell activation mediated by T cell receptors (TCRs) [9,10]. By way of in vivo studies, Kim et al. showed that transplantation of tonsil-derived-MSCs into imiquimodinduced psoriatic skin inflammation in mice significantly abrogated disease symptoms, mainly by blunting the Th17 response in a PD-L1-dependent manner [11]. Furthermore, Nie et al. demonstrated that human MSCs alleviated the pulmonary fibrosis by the PD-L1/PD1 signaling pathways [12]. In the bleomycin-induced pulmonary fibrosis model, abnormal PD1 expression in circulating T cells and lung tissues of patients with pulmonary fibrosis was observed [12].
Biologically, MSC-associated PD-L1 may interact with its receptor PD1 on the surface of T cells, leading to the exhaustion of T cells pathologically involved in the development of aGvHD. In aGvHD patients, Erkers et al. reported that decidual stromal cells have cell-cell contact in order to suppress lymphocytes and blocking PD-L1 impaired their antiproliferative ability [13]. In liver transplant recipients diagnosed with aGvHD, higher PD1 expression was observed on donor CD4 and CD8 T cells. Blocking PD-L1 on the host-derived cells significantly enhanced alloreactivity by CD8 T cells in vitro [14]. This finding suggests that augmenting PD1/PD-L1 pathway may be a therapeutic strategy to control graft-versus-hostreactive T cells. In a mouse model of aGVHD, Tang et al. reported that systemic overexpression of PD-L1 inhibits the donor T cells activation, effector memory status, as well as Th1 and Th17 cells' responses in vivo [15]. Furthermore, they demonstrated that PD-L1 Ig treatment led to decreased T cells proliferation, promoted apoptosis, and reduced pro-inflammatory cytokine expression by effector T cells in vitro, in the setting of anti-CD3/CD28 stimulation and allogeneic dendritic cell stimulation [15]. These studies suggest that MSCs provide the regulatory checkpoints necessary to target pathological CD4+ T cells during stem cell transplantation treatments.
Extracellular vesicles (EVs) include a heterogeneous group of lipid bilayer membranous structures released from their host cells including mesenchymal stem cells (MSCs). Small EVs (sEVs, also referred to as exosomes) usually have an average size between 30-150 nm [16]. It has been suggested that exosomes are effective at delivering many types of bioactive molecules, including mRNAs, miRNAs, lncRNAs, lipids, and proteins to a variety of target cells and that they can actively regulate the target cells even in organs distant from the cell of origin [17].
Meanwhile, more and more studies also suggest that tumor exosomal PD-L1 contributes to immunosuppression and is a valuable target for therapy and diagnosis. For instance, Cheng et al. reported that metastatic melanomaderived exosomal PD-L1 is a rational predictor for patients and suppressed the function of CD8 T cells related to tumor growth [18]. In non-small cell lung cancer, Kim et al. reported that exosomal PD-L1 promotes tumor growth through immune escape [19]. Very similar observations were obtained in the studies of head and neck cancer [20] and gastric cancer [21]. Poggio et al. suggested that suppression of tumor exosomal PD-L1 may induce complex anti-tumor immunity and memory [22]. Therefore, both WJMSC-derived and tumor-associated exosomes which carry PD-L1 demonstrate the specific capability to target T cells through intervening with signaling pathways related to TCRs' functions.
Like tumor-derived exosomes, recent studies have suggested that that MSC-derived exosomes can affect CD4+ and CD8+ T cells in the aGvHD patients. Lai et al. reported that MSC exosomes effectively prolonged the survival of chronical GVHD mice and diminished the clinical and pathological scores [23]. They observed that activated CD4 T cells and their infiltration into the lung were reduced in these animals. Wang et al. reported that WJMSC-EVs alleviated the in vivo manifestations of aGVHD and the associated histologic changes and significantly reduced the mortality of the recipient mice [21]. They also found that recipients treated with hUC-MSC-EVs had significantly lower frequencies and absolute numbers of CD3+CD8+ T cells. Similar findings were obtained in aGvHD mouse model. Fujii et al. reported that the systemic infusion of human bone marrow MSC EVs prolonged the survival of mice with aGVHD and reduced the pathologic damage in multiple GVHD-targeted organs [24]. They reported that both CD4+ and CD8+ T cells were suppressed in EVtreated GVHD mice. As shown in Figure 1, we have recently reported that WJMSC sEVs suppress activated CD4+ T cells that were stimulated with anti-CD3/CD28 beads [1]. This observation suggests that WJMSC sEVs intervened on the CD4+ T cell activation and we have shown that multiple inhibitory checkpoints exist on WJMSC sEVs.
We identified PD-L1 enriched on this WJMSC sEVs as one of the key regulators for suppressing T cells [1]. We observed that exo-PD-L1 from WJMSCs demonstrated some unique features. First, exo-PD-L1 is a membranedbound protein even though exosomes themselves are very diffusible; second, PD-L1 is highly enriched on the sEVs and preserves its bioactive glycosylation as previously reported [25]; and third, exo-PD-L1 is inducible through licensing WJMSCs by proinflammatory cytokines, such as interferon-gamma (IFNγ). These features support that exosomes may provide a functional form of PD-L1 that can be isolated to create a cell-free therapy. Exosomal PD-L1 is also likely superior to a recombinant PD-L1 for its ability to persist in vivo, partially due to co-expression of CD47 on sEVs [26,27], and home to various inflammatory sites, although the mechanism behind this ability is poorly characterized [26]. Consistent with the above findings from tumor exosomes, we found that WJMSC-derived sEVs also rely on PD-L1 to enhance the exhaustion of activated CD4+ T cells. Both pharmaceutic blocking of PD-L1 protein and genetic disrupting of the PD-L1 gene can dramatically impair WJMSC sEV-mediated inhibition of CD4+ T cells. More importantly, WJMSC exo-PD-L1 seems more efficient to block the activation of CD4+ T cells than both soluble PD-L1 or cell surface PD-L1. These observations support the concept that therapeutic WJMSCs provide other exosome-derived checkpoint signals which synergize with the exo-PD-L1 to regulate the immune activities in aGvHD patients.
Kordelas et al. reported that MSC EVs reduced the proinflammatory cytokine response of patient's PBMCs and the clinical GvHD symptoms during the course of MSC EVs therapy [28]. In support, we observed a rapid increase of peripheral plasma exo-PD-L1 in aGvHD patients shortly after infusing with therapeutic WJMSCs [1]. These J Cancer Immunol. 2021 Volume 3, Issue 2 100 findings suggest that increased levels of exo-PD-L1 are correlated with the therapeutic infusion of WJMSCs. In aGvHD patients, we hypothesize that licensing WJMSCs is a necessary step to induce high levels of exo-PD-L1, further contributing to the "pool" of exo-PD-L1 in patient blood ( Figure 1). However, we also noticed that the distribution of exo-PD-L1 in patients' circulation system is completely dependent on the subsequent cell uptake from other blood cells. Thus, our group's study as well as other groups' studies provide strong evidence to support that exo-PD-L1 is essential for WJMSC-mediated therapy of aGvHD in acute AML patients after HSCT.
In summary, exosome-associated proteins derived from therapeutic WJMSCs can provide important regulatory signals to impact adaptive immune activities. Further, identifying and characterizing these exosomal immune regulators will help us gain deeper insights into the therapeutic value of WJMSCs and WJMSCderived exosomes in the treatment of complex human immune diseases. Finding the balance between inhibiting alloreactive T-cells mediating aGvHD without impairing the GvL effect and tumor elimination is a complex issue when it comes to leveraging immunosuppressive treatments after HSCT. However, we believe it is possible to find this balance through the proper timing and dosing of the WJMSC-associated sEVs. Critically, these small nano-size particles may be very helpful in developing novel cell-free therapies for many patient populations.