Chimerism refers to the stable persistence of a group of cells in another genetically distinct individual. In microchimerism (MC) the circulating cell population is below 5%. MC can be observed after non-leukoreduced and leukoreduced blood product transfusion and in transplant recipients, twins, and pregnant women [29, 30]. The lifespan of peripheral blood cells varies from hours to around 100 days [31]. Transfused leukocytes are expected to be completely cleared by alloimmune recognition and/or natural cell senescence. However, Lee et al. observed a transient proliferation of donor white blood cells in canine and human recipients circulation 3–5 days after unmanipulated packed RBC transfusion [32, 33]. Transfusion-associated microchimerism (TA-MC) was observed in 45% of severe trauma surgery patients and sometimes lasted for years [33, 34]. TA-MC may initially result from the proliferation of passenger leukocytes and, in the long term, from the differentiation of donor peripheral blood stem cells (PBSC), which can be present in the peripheral circulation at any time [35].

Like in trauma surgery, transplantation recipients may also suffer from peri-operative fluid loss, prolonged involuntary thermoregulation under anesthesia and/or extracorporeal life support, ischemia-reperfusion injury, and infection risks due to exposure to the environment (e.g., pathogens in recipient bronchus could contaminate the chest cavity during LTx). Their immune system is over-stressed, which creates a favourable environment for TA-MC to exist and continue to regulate recipient’s immune system long-term after transplantation.

In a mouse allogenic femur transplant model, Bingaman et al. found that hematolymphopoietic chimerism (due to bone marrow transplantation) could lead to long-term donor-specific hyporesponsiveness [36]. In 1999, Spitzer et al. conducted a histocompatibility leukocyte antigen (HLA)-matched bone marrow and kidney transplant on a preoperatively induced female patient with multiple myeloma and end-stage renal disease. The low dose of cyclosporine monotherapy was completely withdrawn on Day 73. Renal function remained stable, with no evidence of acute or chronic rejection, and the patient survived over 5 years after transplantation [15, 16].

Starzl et al. proposed a two-way paradigm to explain how MC can induce tolerance. The outcome of transplantation is influenced by both host-versus-graft (HvG) and graft-versus-host (GvH) immune reactions, regulated by the migration and localization of the respective immunogenic leukocytes. If the donor antigen could primarily bypass or secondarily avoid collection by recipient lymphoid tissue, where the passenger leukocytes preferentially migrate to, the immune response could not be induced and the recipient could remain ignorant of the graft existence. This process is mediated by multiple cytokine and receptor pathways. For example, the persistence of both immune reactions could trigger mutual clonal exhaustion-deletion through FasL and tumor necrosis factor (TNF) pathways, which would be crucial for tolerance induction [37–40].

Following blood transfusion, a large amount of iron released from damaged red blood cells and present in plasma is phagocytized by monocytes. The iron homeostasis relevant intra-graft gene expression can predict tolerance in liver transplantation. A higher serum level of hepcidin and ferritin and increased hepatocyte iron deposition were found in operationally tolerant liver transplant recipients [41]. The extracellular iron levels and the balance among ferritin generation and secretion and the primary form of ferrous iron storage play a crucial role in monocyte function [42, 43].

Excessive cellular iron level can impair transcription factor regulation, such as reducing activation of nuclear factor interleukin 6 (NF-IL6) and hypoxia-inducible factor-1 α (HIF-1α) and inhibiting phosphorylation of signal transducer and activator of transcription 1 (STAT1) [44]. NF-IL6 plays a central role in cytokine and iron-mediated regulation of nitric oxide synthase (NOS) expression. Reduced NF-IL6 can downregulate the expression of inflammatory cytokines (such as IL-1 and TNF) and granulocyte colony-stimulating factor (G-CSF) in mature macrophages [45, 46]. The absence of HIF-1α results in ATP level decreasing in macrophages further reducing phagocytosis and migration. HIF-1α deletion can cause a reduction of inflammatory cytokines such as IL-1β, IL-6, IL-10, TNF, and IFN-γ in macrophage and/or dendritic cells (DCs). HIF-1α is also essential for pro-inflammatory M1-type macrophage polarization and maturation of DCs [47–49]. STAT1 can regulate the number and phenotype of macrophages [50]. The STAT1 deficiency can abolish STAT1-dependent cellular response to both INF-α and -γ resulting in immunodeficiency [51, 52].

Iron overload can also directly affect the expansion and function of effector T cells. In patients with hereditary hemochromatosis, where iron overload is a prominent feature, the proliferative capacity, numbers, and activity of cytotoxic T cells (CD8⁺CD28⁺) were decreased, while the number of CD8⁺CD28⁻ T cells was increased [53–55]. Abundant CD8⁺CD28⁻ T cell numbers were associated with better graft function and reduced rejection by inhibiting antigen-presenting cell (APC) allo-stimulatory capacity in liver transplant patients [56]. Consequently, iron overload impairs phagocytosis and antigen presentation leading to decreased activation of effector T cells [57, 58].

Additionally, soluble MHC class I and II molecules (sMHC-I and sMHC-II) are distinctive components in DSBT(WB) compared to regularly stored red blood cell transfusions. sMHC molecule carries donor tolerogenic peptides that can bind to recipient T cells through T cell receptors (TCR). sMHC-TCR binding competing with recognition by APCs results in receptor blockade and apoptosis of recipient T cells due to the absence of co-stimulatory molecules [59–61]. Calne raised a concept of “the liver effect” to describe the immunosuppressive effect of liver transplantation on the other allograft [62]. Graeb et al. further showed in a rat [ACI(RT1^a) → Lewis (RT1^I] model that donor liver-produced sMHC could suppress immune response in recipients and protect heart allograft from rejection [63]. The balance between sMHC-I and II has also been reported to be linked to immune homeostasis [64, 65].

Taken altogether, DSBT(WB) has the potential to block both the antigen presentation of mononuclear phagocytic cells and the activation of effector T cells. Transplantation in such an environment could therefore facilitate graft acceptance [66].

Tregs account for 5%–10% of the T cell population in peripheral blood and contribute to immune homeostasis by regulating innate and adaptive immune responses [67]. Particularly CD4⁺ Tregs expressing forkhead box protein 3 (Foxp3) have been shown to regulate alloimmune response after transplantation. In both animal studies and clinical research, it was found that Tregs interact with other effector T cells through inhibitory cytokines, cytotoxicity, and direct contact [68–71]. Compared to deleukocyted blood product, donor-Tregs could be transfused into recipients through DSBT(WB) and survive due to TA-MC, further mediating immune response after transplantation. Furthermore, it has been observed in a rat model [ACI (RT1.A^aB^b) → Lewis (RT1.A^lB^l)] that DSBT can also induce active expansion of CD4⁺ T cells and donor-specific Tregs in the DSBT recipient’s spleen by indirect allorecognition via residents DCs [72].

In the skin transplant model, CBA/Ca (H-2^k) mice received weekly intravenous transfusions of 0.25 mL whole blood from donor C57BL/10 (H-2^b) for five cycles. CD4⁺CD25⁺ T cells from the mesenteric lymph nodes and spleens of transfused mice, collected 1 week after the final transfusion, were co-transferred with naïve CD45RB^hi cells (effector cells) into CBA-Rag deficient recipient mice 1 day before skin transplantation. This protocol induced long-term tolerance, with 100% skin allograft survival after 100 days without immunosuppression [73]. Similar outcomes were observed in rat heart and intestinal transplants [RA (RT1^p-RT1A^c:B/D^c) → PVG (RT1^c-RT1A^u:B/D^l)], where DSBT induced the development of CD4⁺CD45RC⁻ Tregs in recipients from post-transplant Day 5. These Tregs were highly effective in transferring donor-specific tolerance, as confirmed by adoptive cell transfer experiments. In addition to DSBT, the generation of Tregs requires the presence of graft, thymus, and spleen. These Tregs can be found in secondary lymphoid tissues and in the graft itself, suggesting a local protective effect [73–75].

Human transplant studies also support the role of Tregs in graft acceptance. In Leuven Immunomodulatory Protocol-treated intestinal transplantation recipients, a high level (1.8%) of circulating CD4⁺CD45RA⁻Foxp3^hi memory Tregs was detected in the graft, correlating with long-term reduction of rejection [27]. Furthermore, the Foxp3^hi Tregs subset was associated with improved outcomes graft and patient survival after kidney transplantation in a cohort of donor-specific hypo-responders [76–78].

Secretion of pro-inflammatory cytokines (IL-2, IL-12, TNF-α, INF-γ, etc.) by type 1 CD4⁺ T helper cells (Th1) promotes differentiation of cytotoxic CD8⁺ T cells, natural killer (NK) cells and macrophages which are associated with acute rejection. Conversely, the function of type 2 CD4⁺ T helper cells (Th2) is more complex and depends on cellular targets, timing, and Th2 cytokine-dependent Tregs. Th2 produce cytokines such as IL-4, IL-6, IL-10 and IL-13 with both pro- and anti-inflammatory functions and hereby facilitate Tregs, mediate macrophage polarization to M2 phenotype, and trigger B cells humoral immune response. Th2 are considered to be more linked to chronic rejection [79–84]. The balance of Th1/Th2 is implicated in the level of immune response post-transplantation.

In the rat model of DSBT [RA (RT1^p-RT1A^c:B/D^c) → PVG (RT1^c-RT1A^u:B/D^l)], the Th1/Th2 cytokine profile differed, and a Th2 bias was observed after heart transplantation. On post-transplant day (POD) 5, INF-γ and IL-10 levels in allograft-rejecting rats peaked significantly higher than in DSBT-tolerized rats, whose levels did not peak until POD 9 and POD 30, respectively. IL-4 levels in DSBT-tolerized rats continued to rise until POD 30, while in allograft-rejecting rats, it peaked on POD 5 and was considerably lower than in DSBT-tolerized rats [85]. The mechanism by which DSBT modulates this Th1/Th2 balance remains unclear. A potential explanation is Tregs expansion after/by DSBT [86]. sMHC-I in DSBT may also regulate Th1/Th2 cytokine expression by decreasing IL-2 and granulocyte-macrophage colony-stimulating factor (GM-CSF) while increasing IL-4 and IL-15 [59].

The WOFIE theory is based on the two-way paradigm hypothesis of Calne and Starzl [38, 87–89]. Tolerance, like rejection, is an active immune response that relies on the balance between graft-versus-host (GvH) and host-versus-graft (HvG) reactions. This balance, crucial for graft survival, is influenced by the graft’s immunogenicity, the recipient’s immune system, and donor immune cells, resulting in graft rejection, tolerance, or GvHD. Despite the potential for HLA sensitization, DSBT does not provoke a strong or sustained immune response. Instead, DSBT creates a window of opportunity for immune engagement by promoting a balanced interaction between graft-versus-host (GvH) and host-versus-graft (HvG) responses.

Early after the allograft is exposed to the recipient’s immune system, the GvH response is higher and HvG is lower in recipients who have received DSBT compared to those who have not. However, this increased GvH response remains balanced with decreased HvG, facilitating the development of immunological tolerance and potentially improving graft survival. Non-specific excessive immunosuppression given before and immediately after the transplant can block this naturally occurring well-balanced tolerogenic response. On the longer-term lower levels of maintenance immunosuppression can prevent an overactive HvG response, thereby maintaining the balance and tolerogenic interaction (Figure 4).

Calne et al. first introduced the concept of WOFIE in porcine renal transplant in 1994, administering irradiated leukocytes from the donor spleen to the recipient 6 h after transplantation, and creating a cyclosporine-free window of 48 h [88]. In 1998 they replaced leukocytes with DSBT on the day of transplantation and prolonged the cyclosporine-free window to 96 h in rhesus monkeys kidney transplantation model [87]. In both models, this limited immunosuppression strategy was proved to be effective in improving graft acceptance. In rat intestinal transplantation [RA (RT1^p-RT1A^c:B/D^c) → PVG (RT1^c-RT1A^u:B/D^l)], Pirenne et al. demonstrated that DSBT-induced tolerance could be disrupted by high doses of methylprednisolone [90]. They further observed in rat heart transplantation [RA (RT1^p-RT1A^c:B/D^c) → PVG (RT1^c-RT1A^u:B/D^l)] that a low level of cyclosporine (10 mg/kg) given peri-transplant could lead to tolerance whereas a high level (50 mg/kg) compromised graft acceptance and recipient survival, by blocking the development of T regs. Additionally, administering 10 mg/kg cyclosporine on POD 0–4 failed to induce tolerance, but proved effective when given on POD 5–9 [91].

TRALI is characterized by the onset of new acute lung injury within 6 hours of a blood transfusion, with no identifiable other risk factors. TRALI can occur in all kinds of blood products transfusion but most frequently in products with >60 mL of plasma [98]. The incidence of TRALI is 0.2‰, making it a leading cause of mortality associated with plasma-containing transfusions in the United States [99, 100]. Diagnosing TRALI can be challenging, particularly in distinguishing it from primary graft dysfunction (PGD) following LTx, as both complications present similar symptoms of hypoxemia and bilateral infiltrates on chest X-ray [99, 101]. Despite this similarity, TRALI is strongly linked to blood transfusions, while PGD may occur up to 72 h after LTx.

The exact mechanism of TRALI remains unclear, but it is believed to be triggered by donor leukocyte antibodies present in blood products [102]. Risk factors for TRALI include major surgery within 72 h, active infection, massive transfusion, and cytokine administration, which primes circulating hematopoietic cells before encountering antibodies, thereby increasing the risk of TRALI [99, 103–105]. Currently, there is no published data on TRALI in the DSBT animal model. In our own preclinical experience of DSBT in mice LTx, we did not observe any event of TRALI after iso- or allo-blood transfusion.

Hyperacute rejection rarely occurs after LTx, clinically featuring sudden hypoxemia, widespread pulmonary infiltration, and newly developed pulmonary hypertension within hours after reperfusion [106, 107]. Acute rejection is more common in about 10% of all adult LTx recipients within 1 year posttransplant [108]. In both rejections, preformed and de novo donor-specific antibodies (DSA) are the risk factors [109]. It has been confirmed in a rat model [ACI (RT1.A^aB^b) → Lewis (RT1.A^lB^l)] that DSBT can induce antibody-forming cells to produce DSA in the spleen [72]. Therefore, DSBT may sensitize recipients resulting in increased risks of hyperacute and acute rejection.

However, Ueta et al. found in the rat model that the de novo DSA were not detectable until Day 5 and reached a peak concentration on Day 7. These DSA targeted MHC-I on donor passenger T cells and suppressed acute GvHD [72]. While it is the anti-DQ (HLA-II) DSA that is more often considered associated with antibody-mediated rejection and worse prognosis in LTx [110–112]. Pirenne et al. also reported no hyperacute or acute rejection events in their DSBT-treated intestinal transplant patients [27]. Whether DSBT-induced DSA can sensitize LTx recipients and cause hyperacute and acute rejections remains controversial and should be closely monitored.

TA-GvHD is a rare fatal complication after blood transfusions. It is characterized by pancytopenia and multiple organ failure, likely triggered by the proliferation of donor T cells in the circulation. These cells not only engraft but also attack host tissues, mirroring the pattern of GvHD [113, 114]. While case reports of TA-GvHD have been observed after liver, lung, and kidney transplantation [115–117], the underlying mechanisms remain inadequately explored.

TA-GvHD is particularly problematic in immune immature recipients, such as infants, due to their inability to recognize and eliminate foreign donor cells, coupled with the presence of shared HLA antigens, which are identified as primary risk factors [118–120]. In the DSBT protocol, the fresh whole blood is not irradiated and deleukocyted, which is an effective preventive measure against TA-GvHD [121]. For transplantation patients, partial HLA matching, especially when the donor is homozygous for an HLA haplotype while the recipient is heterozygous, results in the situation that the recipient’s immune system fails to identify and clear the donor-specific leukocytes. In contrast, the donor leukocytes are activated to target the recipient tissue [122]. Transplant recipients could also be immunodeficient due to poor preoperative status and induction immunosuppression during the window of DSBT, which raises the risk of TA-GvHD. Retrospective analysis suggests that the dose of lymphocytes, approximately 10⁷ lymphocytes/kg of recipient weight, correlates with the risk of susceptible TA-GVHD cases [123]. It suggests that the volume of DSBT is not “the more, the better,” and the preoperative evaluation should be more cautious based on the HLA haplotype status.

An animal model of DSBT in LTx could provide essential insight into determining the best timing and dose of DSBT, immune cell differentiation after DSBT and the role of TA-MC in immune regulation. A previously published review has discussed the pros and cons of animal LTx models [124], and we propose that the rodent model stands out as a more suitable choice for DSBT research in LTx, considering factors such as cost, surgical complexity, and availability of analysis techniques.