In the tumor microenvironment of DLBCL, a significant population of immune cells is present, and macrophages plays a pivotal role. M2-type macrophages exhibit a substantial increase within the DLBCL microenvironment, potentially contributing to the establishment of an immune-suppressive milieu, thereby limiting the efficacy of immunotherapeutic approaches. Understanding the proteins and pathways associated with M2-macrophages can facilitate the enhancement of immunotherapy’s effectiveness or the mitigation of the tumor’s resistance to immune attacks. In this study, we utilized public databases to construct a risk model for MDGs in DLBCL. Subsequently, we validated that the expression levels of M2 macrophages-associated gene, HIF1a, in specimens were positively correlated with favorable prognoses in DLBCL.
We have identified that the M2 macrophage-associated genes we screened are primarily related to cytokine pathways. Several studies have pointed out that M2 macrophages can secrete multiple cytokines to participate in the microenvironment regulation of DLBCL. Firstly, M2 macrophages can promote the formation of immune escape in tumors. For example, the overproduction of Interleukin-10 (IL-10), which is secreted by M2 macrophages, can suppress the activity of immune cells, especially T cells and Natural Killer (NK) cells, reducing their attacks on tumors. There are reports suggesting that in muscle-invasive bladder cancer, macrophages secrete IL-10, inhibiting the functions of CD8 + T cells and NK cells, thereby creating an immune escape microenvironment in tumors, and this may be associated with the EGF signaling pathway[12]. Additionally, Transforming Growth Factor-beta (TGF-β), another anti-inflammatory cytokine secreted by M2 macrophages, can inhibit the activity of T cells and reduce their cytotoxicity against tumor cells. There have been reports in B-cell lymphoma that TGF-β can upregulate the expression of CD70, which in turn inhibits the functions of effector memory T cells, and these CD70 high-expressing effector memory T cells are positively correlated with immune checkpoint PD-1 and TIM3 expression[13]. In T-cell lymphoma, IL-4 and IL-13 are increased due to the influence of STAT6, promoting the polarization of M2 macrophages and enhancing tumor immune suppression[14]. In summary, a series of genes related to M2 macrophages are closely associated with cytokines in the microenvironment of DLBCL tumors and patient prognosis.
Our subsequent results also suggest that the high-risk group related to M2 macrophages expresses more CD274 and ENTPD1, implying a potential role of M2 macrophages in immune therapy for DLBCL. In a study of primary cutaneous diffuse large B-cell lymphoma, it was found that all 29 cases expressed PD-L1 in M2 macrophages, suggesting the potential therapeutic value of reprogramming tumor-associated macrophages through anticancer therapies[15] .
Furthermore, M2 macrophages and CD4 T cells have an intricate relationship in regulating immune responses, including in some cancers. Our results showed that patients in the group of low-risk have more CD4 T cells. M2 macrophages can inhibit the activation of CD4 T cells through secreting different kinds of cytokines such as IL-10. In a study of dedifferentiated liposarcoma, immune cell infiltration also predicts clinical outcome, with CD4 T cells related with better prognosis while M2 macrophages are associated with worse outcomes[16]. In addition, CD4 T cell dependent differentiation of M2 macrophages are increased due to the deficiency in interleukin-12p35 regulates angiotensin II[17]. Overall, the relationship between M2 macrophages and CD4 T cell is highly dynamic, the specific molcecular mechanisms need further investigation.
Based on the impact of M2 macrophage-related genes on the prognosis of DLBCL patients, we constructed a LASSO logistic regression model. Using this model, we identified key genes that significantly affect prognosis and used them to construct a risk scoring system, further refining the prognosis classification of patients.
To further validate the risk-related differential genes we discovered, we conducted tissue validation and experimental validation. These validation steps help ensure the reproducibility and biological significance of our findings. Finally, we found that HIF1a is under expressed in the poor prognosis group of DLBCL and is negatively correlated with M2 macrophage content.
HIF1a is activated when cells are exposed to low oxygen conditions (hypoxia), which is always happened in TME and further promoting angiogenesis. However, in our analysis, we observed a negative correlation between HIF1a and M2 macrophage expression. Previous research has revealed that HIF1a can induce the expression of heme oxygenase-1 (HO-1), which acts as a scavenger for reactive oxygen species (ROS). HO-1 inhibits the differentiation of immature Hodgkin lymphoma cells by reducing ROS levels[18]. This study underscores the unique role of ROS in B cell differentiation. Consequently, we hypothesize that HIF1a may be associated with B cell differentiation, although further research and exploration are required to confirm this.
Furthermore, M2 macrophages mitigate the body's inflammatory responses and promote tissue repair by secreting cytokines. Consequently, the oxygen content within tissues may improve, potentially leading to a reduction in the activity of HIF1a. The activation of HIF1a is also subject to regulation by certain inflammatory signaling pathways, such as the Toll-like receptor (TLR) signaling pathway. It has been reported that in fibroblasts, the coupling of ubiquitin to the endoplasmic reticulum degradation domain of Toll-like receptor proteins is a necessary condition for activating HIF1a. The loss of Toll-like receptor function can result in reduced expression of HIF1a, consequently activating the STAT5 pathway[19]. When Toll-like receptors bind to their ligands, they can activate downstream signaling pathways, including NF-κB and MAP kinases, leading to the generation of inflammatory responses. Some researchers have noted that the inhibition of glutamine synthetase can prompt macrophages to polarize from the M2 type to the M1 type, inducing metabolic changes in macrophages. These changes include a reduction in intracellular glutamine, an increase in succinate, and enhanced glucose flux within the glycolytic pathway, ultimately culminating in the activation of HIF1a expression. Therefore, the activation of HIF1a appears to be associated with metabolic alterations induced by GS inhibition and shifts in macrophage functionality[20]. Consequently, the reduced expression of inflammatory signals due to the decreased chronic inflammation caused by M2 macrophages may contribute to the downregulation of HIF1a expression.
The potential impact of the risk model associated with M2 macrophages and changes in HIF1a expression on clinical practice and treatment strategies for DLBCL patients is a subject that warrants further investigation. It's important to note that this study is a retrospective bioinformatics analysis, and the reliability of our risk model requires validation with larger and more diverse datasets. These factors may have implications for the clinical practice and treatment strategies for DLBCL patients.