Background
Acute myeloid leukemia (AML) is the most common form of acute leukemia in adults [
1] and current treatments remain unsatisfactory. Serious infections, resistance to therapy and relapses are the main causes of mortality among patients [
2‐
4]. In particular, the high frequency and severity of infections (especially fungal) before or during chemotherapy are probably due to a severe adaptive immunity dysfunction directly induced by the disease [
5,
6]. This immunosuppressive state may also be responsible for the continuous recurrence of AML and for the failure of immunotherapies [
7], with the exception of allogeneic transplantation [
8].
Although several studies identified different immunosuppressive mechanisms operating in AML [
9‐
13], the precise link between immune alterations, leukemia immune escape and infections has not yet been elucidated. Investigating in this direction, T helper cells seem to be important players with Th17 cells being one of the most intriguing and not fully understood subset so far [
14].
Th17 cells show a pleiotropic role in the inflammatory response, autoimmune disorders and tumors [
15‐
19]. In the latter context, they showed remarkable epigenetic plasticity [
14] and the ability to transdifferentiate into T helper 1-like cells (secreting IFN-γ and showing tumor suppressor activity) [
19‐
21] or T regulatory (Treg)-like cells (secreting IL-10, with immunosuppressive functions and probably tumor promoter activity) [
22‐
25].
Moreover, Th17 cells have the important task of coordinating the immune defense against bacterial and fungal infections and physiologically protect humans against these diseases [
26‐
28]. However, the behavior of Th17 in several tumor types and hematologic malignancies remains to be clarified [
22,
29‐
31]. In particular, different studies on AML have attributed a controversial pathogenetic role and divergent prognostic values to these cells [
31‐
36] but have not succeeded in establishing a link between the reported alterations and the infections to which these patients are subject.
We showed that Th17 cells with a double production of IL-17 and IL-10 were strongly increased in AML patients and that ex vivo patient immune response to an infectious antigen, such as Candida Albicans (C. Albicans), was significantly reduced by Th17. Finally, we found that blasts, co-cultured with CD4+ cells from healthy donors, were able to change the frequency and cytokine profile of T cells, and in particular of Th17, in a similar manner to that observed in patients.
All of the above data support the hypothesis that the increase in IL-10+ Th17 cells in AML is a mechanism developed by the disease to create an immunosuppression state which, given the stem-like features and long life of Th17 cells [
14,
37], may be durable and ultimately favor infections, protecting leukemia cells from immune control.
Methods
Blood samples and PBMCs collection
After obtaining the patient’s informed consent and the approval of the local ethics committee, in accordance with the Declaration of Helsinki, samples of peripheral blood (15–20 ml) were collected from 30 newly diagnosed AML patients before any treatment was started and from 30 age-matched (±10 years) healthy volunteers (HV). AML patients were diagnosed according to the French American-British (FAB) classification system [
38]. Patient and HV characteristics are reported in Table
1. Blood samples were collected in sterile EDTA tubes and mononuclear cells (PBMCs) were separated by density gradient centrifugation using Lymphosep (Biowest) and frozen in 90% heat inactivated fetal bovine serum (FBS) (PAA) and 10% dimethylsulfoxide (Sigma Aldrich).
Table 1
Patient and HV characteristics
No. patients | 30 | 30 |
Gender | | |
Male | 15 (50%) | 14 (47%) |
Female | 15 (50%) | 16 (53%) |
Median age, years (range) | 63 (38–87) | 68 (35–85) |
Subtype according to FAB classification |
M0–M1 | | 6 (20%) |
M2 | | 9 (30%) |
M4 | | 4 (13.3%) |
M5 | | 9 (30%) |
M6 | | 1 (3.3%) |
M7 | | 1 (3.3%) |
Karyotype |
Normal | | 14 (47%) |
Undefined | | 4 (13%) |
Complex | | 3 (10%) |
Trisomy chr 8 | | 3 (10%) |
t(8–21) | | 2 (7%) |
Tetrasomy chr 21 | | 1 (3%) |
Inv chr 3 | | 1 (3%) |
Del chr 7 | | 1 (3%) |
Del chr 20 | | 1 (3%) |
Molecular mutations |
FLT3/NPM wt | | 17 (57%) |
FLT3 ITD | | 4 (13%) |
FLT3 mut NPM mut | | 3 (10%) |
FLT3 mut NPM wt | | 1 (3%) |
FLT3 wt NPM mut | | 1 (3%) |
AML-ETO | | 2 (7%) |
Undefined | | 2 (7%) |
CD4+ cell isolation and culture
In order to avoid contamination by CD4+ cells that release IL-17, such as macrophages [
37], PBMCs were thawed and human CD4+ T cells were isolated by negative depletion of CD8+, CD14+, CD15+, CD16+, CD19+, CD36+, CD56+, CD123+, TCR y/δ and CD235a+, using the CD4+ T cell isolation kit (Miltenyi Biotec). In this way also AML blasts, where present, were included in the subsequent analysis. Cells were cultured in RPMI 1640 medium (PAA) supplemented with 10% heat inactivated FBS, 2 mM
l-glutamine (Euroclone), penicillin (100 U/ml) and streptomycin (100 μg/ml) (PAA). CD4+ cells were primed for 24 h at 37°C with IL-6 (30 ng/ml) (Miltenyi Biotec) or TGF-β (10 ng/ml) (Abcam) or a combination of IL-6 and TGF-β. T cells were then incubated for 5 h at 37°C with phorbol 12-myristate-13-acetate (PMA, 50 ng/ml) and ionomycin (1 μg/ml) (Invitrogen) in the presence of GolgiStop Protein Transport Inhibitor (BD Pharmingen). An unstimulated control prepared by incubating CD4+ cells with GolgiStop Protein Transport Inhibitor only was included for each experiment.
Immunophenotypic analysis of T cells
After stimulation, cells were fixed and permeabilized with Cytofix/Cytoperm (BD Biosciences) then immunophenotyped for intracellular IFN-γ, IL-4 and IL-17A expression using the human TH1/TH2/TH17 phenotyping kit (BD Pharmingen) following the manufacturer’s protocol. For Treg analysis, naïve PBMCs were stained with anti-human FITC CD4 (0.6 μg/ml, clone SK3; BD Biosciences) and anti-human APC-Cy7 CD25 (2.5 μg/ml, clone M-A251; BD Biosciences) for 10 min at 4°C in the dark. After incubation, cells were fixed and permeabilized and then stained with anti-human APC FoxP3 (1:11, clone 3G3; Miltenyi Biotec) for 30 min at 4°C in the dark. Appropriate isotype controls were included for each sample.
Cytokine secretion analysis
Stimulated CD4+ cells were washed with cold PBS containing 0.5% (v/v) bovine serum albumin (BSA) (Sigma Aldrich) and 2 mM of EDTA and analyzed using human IL-17 and IL-10 secretion assay—detection kits (Miltenyi Biotec). Briefly, cells were stained with IL-17 and IL-10 catch reagents for 5 min on ice, incubated for 45 min at 37°C to allow cytokine secretion and then with anti-human PE IL-17A, anti-human APC IL-10 and anti-human FITC CD4 for 10 min on ice, according to the manufacturer’s instructions. Samples were washed and suspended for flow cytometric analysis.
CD33+ cells isolation
Circulating CD33+ cells were magnetically isolated from AML PBMCs in two steps: first, CD4+ and blast cells were negatively purified using the T cell isolation kit, as already described; subsequently, CD33+ cells were purified with CD33 MicroBeads kit (Miltenyi Biotec) following the manufacturer’s instructions.
Direct and indirect allogeneic co-cultures
For direct co-cultures, CD33+ cells isolated from 15 AML patients and allogeneic CD4+ T cells obtained from 15 HV as previously reported were co-seeded in 1:1, 1:5 and 1:10 ratios. For indirect co-cultures, purified CD4+ cells were seeded in the bottom part of the 6-well plates of transwell cell culture system (pore size 0.4 μm; Costar Corp.), whereas CD33+ cells were seeded in the corresponding transwell cell culture inserts. In addition, each cell type was seeded individually in 6-well plates for single culture as control. All samples were cultured in complete medium and stimulated as previously described. At the end of stimulation, T cell immunophenotypic and cytokine secretion analysis was performed.
T cell activation with C. Albicans and isolation of IL-17-secreting cells
CD4+ cells (2.5 × 106) were stimulated for 24 h at 37°C with 1 μg/ml of C. Albicans peptides (JPT, Berlin, Germany). During the last 5 h of incubation, cells were maintained in the presence of GolgiStop Protein Transport Inhibitor (BD Pharmingen). Samples were centrifuged at 4°C, incubated with 2 mM of EDTA in PBS for 10 min at 37°C, washed with 0.5% BSA and 0.1% sodium azide in PBS. Cells were then depleted of IL-17-secreting cells using the IL-17 Secretion Assay—Cell Enrichment and Detection Kit (Miltenyi Biotec). The IL-17 specific catch reagent was attached to the cell surface as previously described, after which cells were labeled with anti-human PE IL-17A and stained with anti-PE microbeads. IL-17-secreting cells were separated through two consecutive column runs, according to the manufacturer’s instructions. Negative fraction was cultured for a further 24 h in complete medium supplemented with 1 μg/ml of C. Albicans peptides and then analyzed for intracellular IFN-γ expression using the human TH1/TH2/TH17 phenotyping kit (BD Pharmingen). A sample stimulated with C. Albicans for 48 h without depletion of IL-17-secreting cells was added as control.
Flow cytometry
Flow cytometric analysis were performed using a FACSCanto flow cytometer (Becton–Dickinson) equipped with 488 nm (blue) and 633 (red) lasers and 50,000 events were recorded for each sample. The acquisition and analysis gates were set on lymphocytes based on forward (FSC) and side scatter (SSC) properties of cells. FSC and SSC were set in a linear scale. For more extensive analysis, gates were set on CD4+ T cell subsets. Flow cytometry data were analyzed with Diva Software (Becton–Dickinson).
Statistical analysis
Data were summarized by descriptive statistics (mean ± standard deviation for continued variables and frequency and percentage for categorical variables). Statistical analyses were carried out using the paired and unpaired two-tailed Student’s t tests and confirmed with the non parametric Wilcoxon signed-rank test. P values <0.05 were considered as significant.
Discussion
AML patients at the onset of disease and during chemotherapy are at high risk of severe and potentially fatal infections [
2‐
4], but such conditions cannot be attributed to neutropenia alone. Indeed, concomitant reduced immune surveillance [
9,
10,
12,
13], favors this infectious trend, worsens prognosis and limits therapeutic possibilities [
3]. Multiple mechanisms of immunosuppression have been identified including indoleamine 2, 3-dioxygenase [
9] and CD200 glycoprotein overexpression [
10], an enhanced Treg activity [
11], an impaired dendritic cell maturation [
12] and PD1- PDL1 axis alteration [
13].
More recently, another immunosuppressive mechanism was described, attributing defective immunological synapse formation to T cells [
6]. All these mechanisms cooperate to suppress immune control on leukemia cells and infections, and also reduce the effect of vaccination or other adoptive T cell transfer strategies [
7]. Nevertheless, the immune system may also be effective in controlling AML, as occurs in hematopoietic stem cell transplantation [
8], and exploring additional ways to use such weapon could have a strong impact on the prognosis of these patients.
To our knowledge no convincing correlation has yet been found between immunosuppression, or specific T cell dysfunction, in AML and the infections to which AML patients are susceptible. Th17 cells, a particular subset of CD4+ cells and their respective cytokines, play a pivotal role in the inflammatory response and autoimmune diseases [
15‐
18] and also direct the defense against bacterial and fungal infections of the gastrointestinal tract, skin, airways and lungs [
26‐
28]. However, their function in several tumor types is controversial [
19,
21,
22], and their involvement in hematological malignancies, in particular AML, remains to be defined [
29‐
36]. Although our results confirmed previously published data [
31,
33‐
36] showing statistically higher Th17 cell percentages in the peripheral blood of newly diagnosed AML patients compared to HV, the alteration in our study was observed together with a significant reduction in Th1 and Th2 frequencies. Moreover, a substantial increase in Tregs was observed, as previously reported by Szczepanski et al. [
11].
Several studies hypothesized a role for Th17 cells in the pathogenesis of AML, but conflicting data on the different prognostic significance assigned to these cells [
31‐
33] suggest an incomplete understanding of the mechanisms involved. In our opinion, the clinical presentation of AML patients, frequently affected by severe fungal or bacterial infections, is the most important event to be taken into account together with the increased percentage of Th17 cells [
31,
33,
36], as confirmed in our experiments. Indeed, these two events are clearly conflicting, given the physiological role of defense of Th17 cells. For this reason, we also performed a more in-depth investigation into the ability of Th17 cells to produce IL-17 simultaneously with other cytokines, focusing our research on IL-10 [
23] and IFN-γ [
21]. A simultaneous production of these cytokines has already been demonstrated in Th17 [
21‐
23,
25,
44], in line with their plasticity [
14], and may also be a sign of their epigenetic transdifferentiation into other T cell types, such as Tregs [
22,
23], which, themselves, may differentiate into FOXP3+ IL-17A cells [
44] or Th17 Th1-like cells, secreting IFN-γ [
20,
21].
We observed a significantly higher increase in the frequency of CD4+ IL-17A+/IL-10+ secreting cells in AML patients than in HV, whereas the percentage of IL-17A+/IFN-γ+ cells remained unchanged. Our results thus suggest that the substantial imbalance between IL-17/IL-10-producing cells (the involvement of FoxP3+IL-17A+IL-10+ cells cannot be excluded) and IL-17A/IFN-γ-producing cells, together with a reduced frequency in Th1 and Th2 cells, may act as an additional immunosuppressive factor in these patients, altering the physiological role of Th17, contributing to the infections and probably promoting leukemia escape. Furthermore, as Th17 cells are long-lived cells with a stem-like molecular signature [
37], their immunosuppressive capacity in leukemia may be powerful and more durable.
Moreover, we demonstrated that the immune response of CD4+ cells isolated from patients was strongly reduced against an infective antigen of fungal origin and, notably, that the selective depletion of Th17 cells from the culture, led to a restoration of IFN-γ production.
To investigate the role of circulating leukemic blasts in the observed alteration, we selected CD33+ cells after depletion of myeloid differentiated cells. All the changes observed in Th17 were induced in vitro by CD33+ leukemic cells, as confirmed by direct and indirect co-cultures of healthy CD4+ cells and AML peripheral blasts. Given that in both co-culture the intensity of T alterations were similar, we hypothesized that leukemic cells action was mediated by soluble factors.
Moreover, patient T cells, depleted of CD33+ blasts, regained the capacity to produce levels of IFN- γ and IL-4 similar to those of HV and showed a decreased ability to simultaneously produce IL-17 and IL-10. Therefore all of the above data suggested the involvement of blasts also in maintaining the immunosuppressive state in AML patients.
Authors’ contributions
GM and SC designed the research study, analyzed the data and wrote the paper; GM, SDM, FF and SC performed the development of
methodology; CP, VG, DC, MC, MBG, AL, SR, PPF, PM, PS and MT performed the acquisition of data; GM, SDM, RN, FF and SC analyzed and interpreted the data; GM, SDM, MG, GM, WZ, DA and SC wrote, reviewed, and/or revised the manuscript; SDM, RN, FF and SC administrative the technical, or material support; GM and SC performed the study supervision. All authors read and approved the final manuscript.