Neutrophils are important players in cancer-related inflammation and contribute to tumor spread and disease progression. The role of neutrophils in the progression of CLL has already been underlined by Gatjen et al. The depletion of Ly6G
+ cells delayed leukemia cell proliferation in the mouse Eµ-TCL1 model [
14]. In these studies, neutrophils isolated from the bone marrow showed no changes in ROS production or in phenotype [
12,
14]. Some differences were observed in cells isolated from blood or spleens, where they establish the leukemia microenvironment. On this basis, we adopted the assumptions of our work to study the phenotype and function of neutrophils in the spleen – the site where the crosstalk between neutrophils and leukemic cells occurs. Nevertheless, we did not deplete the Ly6G
+ cells since the results showing the depletion of populations may raise some controversies. Recently, Boivin et al. showed that neutrophils remaining after anti-Ly6G treatment are newly derived from the bone marrow, and these new neutrophils have lower Ly6G membrane expression, leading to a diminution in the target level for anti-Ly6G Abs [
34]. To date, numerous studies have highlighted the plasticity of neutrophils in different pathological states [
35‐
37] and described various approaches enabling the reprogramming of neutrophil functions [
38‐
40]. Importantly, the ability of tumor cells to change neutrophil phenotypes and functions has been broadly reported in studies on the solid tumor microenvironment. Nevertheless, the protumor phenotype of neutrophils is not fully defined, as these cells are challenging to study due to their short lifespan and the inability to genetically manipulate, cryopreserve or expand them in vitro [
41]. Moreover, the availability of data concerning neutrophils in hematological diseases is limited. Therefore, in this study, we aimed to define the neutrophil phenotype of TCL1 leukemia-bearing mice isolated from the leukemia microenvironment. In the TCL1 mouse model, similar to human CLL, we observed a lower number of neutrophils, which is associated with disease progression [
6]. It was already reported by Manukyan et al. that human neutrophils of CLL patients are activated and reveal functional defects that can be related to the clinical course of diseases [
13]. In our studies, neutrophils from TCL1 leukemia-bearing mice also revealed alterations, as we detected a significant decrease in the percentage of MHC-II- and CD80-expressing neutrophils. However, our data showed that the frequencies of CD86-expressing neutrophils decreased at the early stage and increased in the late stage of the disease. The CD86 marker on neutrophils may be modulated by various inflammatory factors, such as IL-4 and TNF-α, which are induced in CLL patients [
42,
43]. Moreover, in contrast to human CLL blood neutrophils, among mouse leukemia-associated neutrophils, we observed high frequencies of CD62L-expressing cells at the late stage of disease [
13]. According to previous studies, CD62L is cleaved from the cell surface during priming or, as presented by Tak et al. [
44], the low expression of CD62L may be associated with a neutrophil subset called “aged”, which is rapidly recruited to the bloodstream in response to acute inflammation [
45]. The results obtained in our study concerning a high percentage of CD62L
high-expressing neutrophils indicate that mouse leukemia-associated neutrophils lost their activity potential, and this effect may be mediated by the factors released by the leukemia microenvironment. Moreover, experiments with TCL1 leukemia-bearing DEREG mice show that Treg depletion reduces the frequencies of CD62L
high expressing cells (meaning increased activation) and induces an antileukemia immune response. Moreover, as previously published by others, Treg can inhibit neutrophil activation and accumulation [
46,
47]. Given the changes we observed in the neutrophil phenotype and the data showing loss of function by neutrophils in CLL patients [
13,
48], a high level of CD62L might be a marker of mouse CLL-associated neutrophil loss of proinflammatory functions. According to our results, mouse leukemia-associated neutrophils exhibit functional defects that manifest in their limited ROS production, decreased capability of phagocytosis, and reduced cytokine production. These results are in agreement with a previously mentioned study by Manukyan et al., who reported functional defects in the circulating neutrophils of CLL patients [
13]. In fact, our data also revealed that the level of azurophilic degranulation was elevated in leukemia-associated neutrophils. Usually, the process of degranulation is tightly controlled; however, dysregulation of content granule release has been reported in cancer diseases. A high level of granule content may be beneficial for tumor development and progression, as shown for many types of solid tumors and acute promyelocytic leukemia [
49]. We assume that the described phenotype provides evidence that TCL1 leukemia-associated neutrophils were polarized into leukemia-supporting cells, similar to observations made by Gatjen et al. [
14]. Moreover, our ex vivo results indicated that factors secreted by TCL1 leukemic cells lead to a decrease in the CD62L
low and an increase in the CD62L
high percentage in the control neutrophil population isolated from healthy animals. This finding may suggest that the CD62L
high neutrophils present in the leukemic spleens are likely modified by a tumor microenvironment.
Tumor-associated neutrophils were consistently reported to reveal immunosuppressive phenotypes. Therefore, we analyzed the neutrophil populations expressing surface molecules known for their immunosuppressive action. Interestingly, flow cytometry revealed that the percentage of TCL1 leukemia-associated neutrophils expressing PD-L1 and ARG-1 was increased at the late stage of disease, whereas the percentage of IDO- and IL4R-expressing cells was higher at the early stage. Our results are in line with the data published by Romano et al. showing that neutrophils in classic Hodgkin lymphoma are immunosuppressive through increased expression of ARG-1, which was higher in patients with advanced stage disease [
50]. Our findings concerning PD-L1-expressing neutrophils are in accordance with currently available data. To date, it has been shown that in some types of solid tumors, increased frequencies of PD-L1-expressing neutrophils are present in the tumor microenvironment. These cells are associated with disease progression and reduced patient survival [
51,
52]. The significant increase in PD-L1 expression in neutrophils after Treg depletion may be a consequence of the release of IFN-γ, the production of which is significantly enhanced as a result of the activation of the antitumor immune response [
22]. Interestingly, IL-10 neutralization also increases PD-L1 expression but only after Treg depletion. These results are consistent with the observation that IL-10 suppression enhances antitumor T cell activity and may explain why the neutralization of IL-10 improves the sensitivity of the TCL1 model to anti-PD-L1 therapy [
53].
Moreover, the results of this study are consistent with the data published by Öztürk et al. showing increased expression of IDO on the surface of leukemia-associated neutrophils in the mouse Eµ-TCL1 model [
54]. Additionally, high expression of IL4Rα in leukemia-associated neutrophils was observed in our study. It was shown that IL-4, through IL4R signaling in neutrophils, suppresses their infiltration and antimicrobial functions [
55] and protects neutrophils from apoptosis [
56]. However, our results obtained from ex vivo experiments suggest that TCL1 leukemic cells induced the expression of IL4Rα on neutrophils, although the adoptive transfer of Treg, was needed to elevate neutrophil IL4Rα in RAG2KO mice in vivo. A previous report showed that Treg can modulate the migration and function of neutrophils through the production of CXCL8 and IL-10 and the modulation of suppressor of cytokine signaling 3 (SOCS 3) in neutrophils [
57,
58]. RNA-seq analysis of the Treg subpopulation from the murine CLL model revealed its unique phenotype characterized by upregulation of multiple factors, including Il10, Ifnγ, and the chemokines Ccl3/4/5 Cxcl13, which likely contribute to the neutrophil leukemia-supporting phenotype. Furthermore, the modification of the tumor microenvironment by induction of the antileukemic immune response and depletion of Treg by DT in DEREG TCL1 leukemia-bearing mice were able to decrease the frequencies of IL4R-expressing neutrophils. Nevertheless, the depletion of IL-10 did not influence the leukemic phenotype of neutrophils in either in vitro or in vivo experiments. Our results also indicated that the production of all tested cytokines, apart from IL-6, was reduced in leukemia-associated neutrophils. This finding additionally highlights the dysfunctionality of neutrophils in CLL. The modification of the tumor microenvironment by induction of the antileukemic immune response and depletion of Treg were not efficient in changing the cytokine profile of leukemia-associated neutrophils.
Although neutrophils isolated from blood or bone marrow have already been studied in the context of the course of CLL, our research focused on the phenotype and function of neutrophils present in the leukemia microenvironment. The obtained results support current data by clearly showing the plasticity phenomenon of neutrophils. We conclude that the change into the immunosuppressive phenotype of neutrophils is induced by B leukemic lymphocytes and Treg, and can be modified again by Treg depletion and the induction of an antileukemic immune response. Interestingly, on the basis of the obtained results, it seems that apart from leukemic cells, Treg can significantly affect the phenotype and functions of neutrophils, which, however, requires further in-depth studies.