Hyperfunction of CD4 CD25 regulatory T cells in de novo acute myeloid leukemia

Regulatory T cells (Tregs), which were originally identified as CD4⁺CD25⁺ T cells, are critical for maintaining immunological self-tolerance in healthy individuals by actively suppressing self-reactive lymphocytes [1]. However, in tumor immunity, Tregs are considered pivotal regulators of immune escape, as they suppress the proliferation and function of immune cells through cell-to-cell contact and inhibitory cytokine production [2]. Previous studies demonstrated that Tregs are increased in many solid tumors [3‐6], as well as hematopoietic malignancies [7‐9]. In addition, an increased Treg frequency is related to poor prognosis [6, 10, 11].

Acute myeloid leukemia (AML) is a hematopoietic malignancy driven by a sequence of somatic mutations in multipotential primitive cells or progenitor cells. Currently, although most AML patients achieve complete remission (CR) after induction chemotherapy, approximately 60% of patients who receive subsequent consolidation chemotherapy still cannot achieve long survival. As immunotherapy is a promising new therapeutic option for hematologic malignancies [12], Tregs can be a novel target. However, little is known regarding the role of Tregs in the immune escape of AML and the possible upstream mechanisms.

As cell-to-cell contact plays an essential role in the immunosuppression of Tregs [13], the accumulation of Tregs in bone marrow (BM) is crucial for AML progression. A previous study indicated that stromal cell-derived factor (SDF)-1α/chemokine (C-X-C motif) receptor (CXCR) 4 signaling plays an important role in regulating Treg levels in BM and peripheral blood (PB) [14]. Blockade of the SDF-1α/CXCR4 axis alone, which is crucial for Treg migration, reduces ovarian tumor growth and peritoneal dissemination and selectively reduces intratumoral Tregs [15]. Therefore, we hypothesize that excessive chemotaxis of Tregs from PB to BM is the reason for Treg enrichment. In addition, regulatory B cells (Bregs) are another negative regulator that supports immunological tolerance by producing interleukin-10 (IL-10), IL-35, and transforming growth factor β (TGF-β), as well as influencing the expansion of T cells [16]. Olkhanud et al. indicated that tumor-evoked Bregs promoted breast cancer metastasis by converting CD4⁺ T cells into Tregs using a mouse 4 T1 breast cancer model [17]. Accordingly, Bregs may play a similar role in inducing Treg conversion in AML immune escape.

To provide an in-depth understanding of Tregs in AML, in the present study, we investigated the proliferation, cytokine production and migration capacity of Tregs in newly diagnosed untreated AML patients.

AML accounts for more than 25% of adult leukemia cases [21]. In addition to the well-known cytogenetic and molecular genetic abnormalities in the pathogenesis of AML, immune escape also plays a significant role. Tregs are considered pivotal regulators of immune escape. As Tregs suppress the proliferation and function of immune cells through cell-to-cell contact [13], the enrichment of Tregs in tumor sites is crucial. In addition, using CD25-specific monoclonal antibodies to deplete Tregs in a variety of different mouse strains promoted the rejection of murine tumor cell lines, including melanoma and leukemia [22‐24]. Previous studies have shown that compared to the frequencies in healthy individuals, the frequencies of Tregs in both PB and BM of AML patients are increased [25, 26]. Nevertheless, little is known about the mechanism of Tregs in the immune escape of AML.

FoxP3 is critical and specific for Treg identification. However, as FoxP3 is expressed intracellularly, and cells to be fixed and ruptured during staining, FoxP3 cannot be used to separate human Tregs for functional studies or in vivo expansion. In addition, CD127 is downregulated in Tregs and is negatively correlated with Foxp3 expression [27]. Therefore, we used CD4⁺CD25⁺CD127^low/− cells as the immunophenotype of Tregs and found that the frequencies of CD4⁺CD25⁺CD127^low/− Tregs in BM and PB were increased in AML patients. We deduced that excessive chemotaxis of Tregs from PB to BM is the reason for this enrichment. The homing of Tregs from PB to BM, which is the target tissue in AML, is crucial for their immunosuppressive function via direct contact with immune cells. The SDF-1α/CXCR4 signaling pathway is critical for Treg trafficking from PB to BM [14]. Moreover, the mechanism of CXCR7, another receptor of SDF-1α, still needs further investigation [19]. Our study demonstrated that the enrichment of Tregs in AML patients was due to excessive migration caused by increased expression of CXCR4 rather than abnormal expression of CXCR7 on Tregs or abnormal secretion of SDF-1α by the hematopoietic microenvironment.

Checkpoint inhibition was taken into consideration as contributing to the immunosuppressive effect of Tregs in AML. A series of inhibitory membrane proteins on the surface of Tregs can exert their immunosuppressive function through cell-to-cell contact [28]. However, in our experiments, differences in the expression of PD-1 on the surface of Tregs [10, 29], were not significantly different between AML and healthy participants, which was consistent with several solid cancers [30‐32]. Therefore, we inferred that the immunosuppressive function of Tregs in AML patients does not rely on the overexpression of PD-1.

In addition, to further investigate the effects of Tregs on other immune cells, we designed in vitro coculture and crisscross coculture experiments. Our study revealed that compared to normal Tregs, AML Tregs were more capable of inhibiting proliferation, promoting apoptosis and suppressing the production of IFN-γ in both normal and AML CD4⁺CD25⁻ T cells. In particular, Tregs from AML patients had the most significant effect on autologous CD4⁺CD25⁻ T cells. Interestingly, the inhibitory effect on CD8⁺ T cells was not significantly different from that of controls. Although IFN- γ production by CD8⁺ T cells showed a similar decreasing trend, the difference was not obvious enough (P > 0.05). In ovarian cancer, a decreased ratio of CD8⁺ T cells to Tregs in tumors is related to poor prognosis, indicating suppression of effector CD8⁺ T cells by Tregs [1]. The effect of Tregs on CD8⁺ T cells in AML could be due to other mechanisms than cell-to-cell contact. Inhibitory cytokines such as TGF-β suppress CD8⁺ T cell proliferation [33], and cytokines could be the manner by which Tregs influence CD8⁺ T cells. In addition, as the function of CD8⁺ T cells relies on the help from CD4⁺ T cells, the immunosuppressive effect could be indirect via CD4⁺CD25⁻ T cells. Crisscross coculture experiments further ruled out the possibility that T cells had decreased resistance to Tregs. Therefore, we suggest that in AML, Treg-mediated immunosuppression of CD4⁺CD25⁻ T cells increases, leading to reduced proliferation, increased apoptosis and impaired secretion of IFN-γ; instead, the resistance of immune T cells to Tregs decreases, which eventually causes immune escape of AML cells.

The in-depth study of AML Tregs raised the question of the possible upstream factors in the initiation of AML Treg abnormalities. Mizoguchi et al. [34]. found a B cell subgroup that secreted IL-10 and inhibited the progression of inflammatory bowel disease, formally suggesting the concept of Bregs. Bregs are involved in the development of various diseases, including autoimmune diseases, infections and tumors. These cells are involved in the regulation of graft-versus-host disease (GVHD), mainly through the secretion of IL-10, TGF-β and other cytokines [35], regulation of T cells, amplification of Tregs [36‐38] and other means of participating in immune regulation. A study showed that the number of Bregs in the PB of liver cancer patients was higher than that in healthy individuals [39]. Olkhanud et al. discovered that Bregs could convert dormant CD4⁺CD25⁻ T cells into CD4⁺CD25⁺Foxp3⁺ Tregs in breast cancer [17]. In our study, the percentage of CD19⁺CD24^highCD38^high Bregs in BM was significantly increased. A coculture experiment proved that AML-induced Bregs robustly converted CD4⁺CD25⁻ T cells to CD4⁺CD25⁺Foxp3⁺ Tregs, while normal Bregs did not. However, as secretion of IL-10 and TGF-β did not show obvious differences between AML and normal Bregs, we deduced that IL-10 and TGF-β were not the reason for the conversion. Little research has been conducted on the impact of Bregs on Tregs. In breast cancer, PD-1⁺ Bregs increase the conversion [40]. Although that study did not use PD-1⁻ Bregs or anti-PD-1 antibodies for comparison, checkpoint inhibitors are still a good entry point for Breg investigations. The conversion effect was most significant when AML Bregs were cocultured with AML CD4⁺CD25⁻ T cells. Therefore, tumor-induced CD4⁺CD25⁻ T cells were more prone to conversion.

In AML bone marrow, the frequencies of Bregs increase and induce the conversion of CD4⁺CD25⁻T cells to CD4⁺CD25⁺Foxp3⁺ Tregs. On the other hand, more PB Tregs home to the BM, which also causes the enrichment of Tregs in the BM. Treg-mediated immunosuppression in immune cells increases, leading to reduced proliferation and increased apoptosis and secretion of IFN-γ, especially for CD4⁺CD25⁻T cells.