CD56neg CD16+ cells represent a distinct mature NK cell subset with altered phenotype and are associated with adverse clinical outcome upon expansion in AML.

doi:10.21203/rs.3.rs-4656048/v1

Download PDF

Article

CD56^negCD16⁺ cells represent a distinct mature NK cell subset with altered phenotype and are associated with adverse clinical outcome upon expansion in AML.

https://doi.org/10.21203/rs.3.rs-4656048/v1

This work is licensed under a CC BY 4.0 License

Version 1

posted

You are reading this latest preprint version

CD56^negCD16⁺Natural Killer (NK) cells have been reported to expand in chronic diseases and acute myeloid leukemia (AML). However, their biological role is still unclear. Using mass cytometry, spectral flow cytometry, bulk RNA-seq and in vitro assays, we characterized CD56^negCD16⁺NK cells that expand in AML. We confirmed that CD56^negCD16⁺NK cells represent a unique NK cell subset coexpressing Eomes and T-bet. CD56^negCD16⁺NK cells could recover CD56 expression in vitro where they displayed unaltered NK cell functions. We previously demonstrated that CD56^negCD16⁺NK cells expansion at diagnosis was associated with adverse clinical outcome in AML. Here, we validated our findings in a validation cohort of N=38 AML patients. AML patients with CD56^negCD16⁺NK cells expansion at diagnosis had decreased overall survival (HR[CI95]=5.5[1.2-24.5], p=0.0251) and relapse-free survival (HR[CI95]=13.1[1.9-87.5], p=0.0079) compared to AML patients without expansion after 36 months follow-up. RNA-seq unveiled that CD56^negCD16⁺NK cells were mature circulating NK cells with functional capacities. Upon expansion, CD56^negCD16⁺NK cells from AML patients showed altered proteomic phenotype, with high expression of TIM-3 and Siglec-7. Taken together, our results suggest that CD56^negCD16⁺NK cells are a relevant target for future NK-cell-based immunotherapies.

Health sciences/Medical research/Translational research

Biological sciences/Cancer/Tumour immunology/Immunosurveillance

NK cells are innate cytotoxic lymphocytes with a pivotal role in the clearance of pathogens and tumor cells(1). They can be divided into three distinct functional subsets based on the surface expressions of CD56 and CD16 : CD56^dim CD16⁺, CD56^dim CD16⁻ and CD56^bright NK cells. Recently, a growing interest in CD56^neg NK cells has arisen. Indeed, this subset present in healthy volunteers, can significantly increase in pathological context and is thought to be a hallmark of chronic NK cell activation(2). Their expansion was first identified in human immunodeficiency virus-1 (HIV-1) infected patients(3) and further characterized by Mavilio and al.(4). Despite no consensus on their maturation stage and functional capacities, there are convincing evidence of their clinical implications in various diseases including HIV-1(5), hepatitis C virus infection(6), malaria(7) or AML as described in our previous study(8).

AML is a haematological malignancy developing within the bone marrow and characterized by a maturation arrest of myeloid precursors and an uncontrolled clonal expansion of poorly differentiated cells, called leukemic blasts. AML is sensitive to NK cell-mediated cytotoxicity, however, leukemic blasts can implement escape mechanisms, resulting in impaired NK cells and poor prognosis(9, 10). We demonstrated that an expansion of CD56^neg NK cells higher than a threshold of 10% at diagnosis was associated with adverse clinical outcome in AML. Interestingly, it persisted in some patients after complete remission.

Although progress has been made in AML therapies over the past decades, it still suffers from poor prognosis in patients aged ≥ 60 with a 5-years survival rate below 15%(11). To restore NK cell cytotoxicity, NK cell-based immunotherapies are currently under investigation(12). The role of CD56 in NK cell cytotoxicity is controversial, however, two recent studies(13, 14) demonstrated its implication in the establishment of immunological synapse, likely to be compromised in CD56^neg NK cells. Molecules like NK-cell engagers(15) could help circumvent the problem and induce synapse formation, triggering NK cells cytotoxicity. Although it is unclear whether CD56^neg NK cells expansion mechanism is common across diseases, it seems more relevant than ever to further investigate this subset, representing a potential therapeutic target.

Hereinafter, we presented evidence that CD56^neg CD16⁺ cells define a distinct NK cell subset. CD56^neg CD16⁺ NK cells could recover CD56 expression and exhibit potent NK cell functionalities in vitro similarly to CD56⁺ NK cells, in a specific culture system. Besides, this subset expressed transcription factors crucial to the maintenance of NK cell identity. Furthermore, we confirmed in a validation cohort of N = 38 AML patients that CD56^neg CD16⁺ expansion above 10% of total NK cells at diagnosis is associated with adverse clinical outcome. Next, we established a transcriptomic signature of CD56^neg CD16⁺ NK cells characterized by hallmarks of mature circulating cytotoxic NK cells, although some transcripts were indicative of impaired cytotoxicity. Finally, we showed that expanded CD56^neg CD16⁺ NK cells exhibit a distinct phenotype characterized by the expression of late maturation markers such as TIM-3 and Siglec-7.

Study Design and Patient Enrolment

New data generated in this study were obtained from frozen peripheral blood mononuclear cells (PBMCs) of patients (N=38) enrolled in the HEMATOBIO cohort (NCT02320656). Blood collection, PBMCs extraction, and storage were performed by the Paoli-Calmettes Tumor Bank, which operates under authorization #AC-2007-33 granted by the French Ministry of Research. Patients with newly diagnosed AML, except acute promyelocytic leukemia, were eligible for study participation. All patients had received standard chemotherapy. The cohort was divided according to the frequency of CD56^negCD16⁺ NK cells with a threshold of 10% as previously defined(8). The clinical information of all patients is summarized in Table 1. Healthy volunteers (HV, N = 16) were recruited from the French Blood Agency (EFS). Written informed consent was obtained from all patients in accordance with the Declaration of Helsinki. The study was reviewed and approved by the Institutional Review Boards of Paoli-Calmettes Institute.

Fluorescence-Activated Cell Sorting (FACS) and RNA extraction

Cryopreserved PBMCs from N=4 AML patients and N=5 age-matched HV were thawed in a 37°C water bath, resuspended in RPMI-1640 (Gibco, Grand Island, NY, USA) supplemented with 10% of heat inactivated foetal bovine serum (FBS) (Gibco), centrifugated for 5 min at 1,500 rpm and incubated in a 37°C, 5% CO2 incubator for 30 min. After resting, cells were filtered on a 30 µm pre-separation filter (Miltenyi), centrifugated and incubated in 1X PBS (Gibco) with the viability marker Live/Dead Fixable Aqua (Thermo Fisher Scientific, Waltham, MA, USA) for 25 min on ice in the dark. After washing in 1X PBS and centrifugation for 5 min at 1,500 rpm twice, cells were incubated with BV605 anti-CD56 (clone NCAM16.2), BV711 anti-CD16 (clone 3G8), BV785 anti-CD45 (cloneHI30), FITC anti-CD13 (clone SJ1D1), FITC anti-CD33 (clone HIM3-4), FITC anti-CD34 (clone 581), PE-CF594 anti-CD3 (clone UCHT1) in Brilliant Stain Buffer (BD Biosciences, Franklin Lakes, NJ, USA) for 30 min on ice in the dark. Before acquisition, cells were passed through a 35 µm cell strainer. Cell sorting of CD56^bright, CD56^dimCD16^-, CD56^dimCD16⁺, and CD56^negCD16⁺ NK cell subsets was performed on a FACS Aria III (BD Biosciences) following the gating strategy outlined in Supplementary Figure 1. All used antibodies are listed in Supplementary Table 1. After cell sorting, total RNA was immediately isolated with a RNeasy Plus Micro Kit (Qiagen, Hilden, Germany) according to manufacturer’s instructions. High-quality RNA was eluted in 14 µl RNase-free water.

Bulk RNA-sequencing and data processing

Samples from patients and HV were respectively pooled, at the same total RNA concentration, according to NK cell subtype prior to sequencing. RNA integrity was assessed on a Fragment Analyzer (Agilent, Santa Clara, CA, USA). Library preparation, RNA Capture Sequencing and were performed by IntegraGen (Evry, France). Briefly, libraries were prepared using NEBNext® Ultra™ II Directional RNA Library Prep Kit for Illumina (NEB, Ipswich, MA, USA) according to supplier recommendations. The RNA was capture with the Twist Human Core Exome (Twist Bioscience, San Francisco, CA, USA) + IntegraGen Custom V2 Enrichment Systems. Sequencing was performed on an Illumina NovaSeq platform and 2 × 100 bp paired-end reads were generated. RNA and sequencing quality control metrics are summarized in Supplementary Table 2.

Spectral flow cytometry, Cell cloning and functional assays, RNA-sequencing analysis, Mass cytometry analysis and Statistical Analysis can be found in Supplementary Materials and Methods.

CD3 ⁻ CD56^neg CD16⁺ cells represent a distinct NK cell subset.

Our team made the first description of CD56^neg CD16⁺ NK cells expansion in AML patients(8). Therefore, we sought to confirm that CD3⁻ CD56^neg CD16⁺ cells identified through mass cytometry belong to NK cell. To this end, we ran Cytosplore^+ HSNE (16) on our derivation cohort data(8) to perform automatic clustering. We embedded 872 796 CD45⁺ cells following the gating strategy outlined in Supplementary Fig. 2 and isolated from peripheral blood of N = 10 HV and N = 22 AML patients. Cytosplore^+ HSNE identified 6 distinct clusters among which NK cells and myeloid/leukemic blasts clusters. Importantly, 88% of CD3⁻ CD56^neg CD16⁺ NK cells clustered among NK cell cluster while 12% clustered among leukemic blasts cluster indicating a strong similarity between CD56^neg CD16⁺ NK cells and conventional NK cells, i.e. CD56^dim CD16⁺,CD56^dim CD16⁻ and CD56^bright NK cells (p < 0.0001) (Fig. 1A & 1B).

Eomes and T-bet transcription factors are crucial to sustain NK cell identity and function(17). Using spectral flow cytometry, we assessed the frequencies of Eomes⁺ and T-bet⁺ cells among CD3⁺ cells, CD56^neg CD16⁺ and conventional NK cells from PBMCs of HV (N = 16) and AML patients (N = 7) with CD56^neg CD16⁺ NK cells expansion at diagnosis (Expanded group). In the HV group, we observed similar frequency of Eomes⁺ cells among CD56^neg CD16⁺, CD56^dim CD16⁻ and CD56^bright NK cells whereas CD3⁺ cells and CD56^dim CD16⁺ NK cells showed lower expression of Eomes compared to CD56^neg CD16⁺, respectively p < 0.0001 and p = 0.0007. In the Expanded group, frequency of Eomes⁺ CD56^neg CD16⁺ cells did not differ from the frequencies of conventional NK cells and was significantly increased compared to CD3⁺ cells, p = 0.0004 (Fig. 1C, left panel). Besides, frequencies of T-bet⁺ CD56^neg CD16⁺ NK cells from both HV and Expanded groups matched the frequencies of conventional NK cells and were significantly increased compared to CD3⁺ cells, respectively p < 0.0001 and p = 0.0007 (Fig. 1C, right panel).

Next, we investigated whether CD56^neg CD16⁺ NK cells express NK cell markers and maintained NK cell killing abilities in vitro. We performed limiting dilution NK cell cloning from whole blood of HV. PBMCs were isolated using density gradient centrifugation and CD56^neg CD16⁺ and CD56⁺ CD16⁺ NK cells were FACS-sorted before being plated at a concentration of 1, 10 or 100 cells/well in RPMI supplemented with 10% of heat inactivated FBS. Irradiated K562 cells were used as feeders. IL-2 (100U/mL), IL-15 (10ng/mL) and IL-21 (5ng/mL) were added every two days and medium was changed. After 24 days of culture, cells were harvested, phenotyped and functionality was assessed (Fig. 1D). We observed that expression levels of CD56 and NKG2A were significantly increased in CD56^neg CD16⁺ NK cells after 24 days of culture, respectively p = 0.0024 and p = 0.0003. Importantly, CD56^neg CD16⁺ NK cells and CD56⁺ CD16⁺ NK cells expressed similar levels of CD56 and NKG2A after 24 days in culture (Fig. 1E). In addition, CD56^neg CD16⁺ and CD56⁺ CD16⁺ NK cells exhibited similar production capacities of IFN-γ and TNF-α after 24 days of culture (Fig. 1F, left and middle panels). Finally, CD56^neg CD16⁺ NK cells expressed higher levels of CD107a/b, p = 0.0057, (Fig. 1F, right panel) and showed similar specific lysis against K562 cells than CD56⁺CD16⁺ NK cells after 24 days of culture (Fig. 1G). Together these data support that CD3⁻ CD56^neg CD16⁺ cells represent a distinct NK cell subset able to recover CD56 expression in vitro where they display unaltered NK cell functions.

Validation cohort confirmed that CD56 ^neg CD16⁺ NK cells expansion at diagnosis is associated with adverse clinical outcome in AML patients.

We demonstrated that CD56^neg CD16⁺ NK cells expansion at diagnosis was associated with adverse clinical outcome in AML(8). To validate our previous results, we included N = 16 HV, N = 31 AML patients without CD56^negCD16⁺ NK cells expansion at diagnosis (Non-Expanded group) and N = 7 patients in the Expanded group within our validation cohort. Patients’ characteristics are summarized in Table 1. PBMCs at diagnosis were analysed using spectral flow cytometry. We previously defined a threshold of 10% of CD56^neg CD16⁺ NK cells among total NK cells to discriminate between patients. As expected, the threshold could optimally classify samples from the validation cohort in the Non-Expanded or Expanded groups (Fig. 2A). Indeed, the Expanded group displayed significant CD56^neg CD16⁺ NK cells expansion compared to the HV and Non-Expanded groups, respectively p = 0.0292 and p < 0.001 (Fig. 2B). Frequencies of CD56^neg CD16⁺ NK cells in the Expanded group ranged from 11.8–75.6% (Supplementary Table 3). Notably, the Expanded group did not show any increase of total NK cells (Fig. 2C, left panel), suggesting that CD56^neg CD16⁺ NK cells expand at the expense of other NK cell subsets. Consistently, the Expanded group had significantly less CD56^dim CD16⁺, CD56^dim CD16⁻ and CD56^bright NK cells than the Non-Expanded (p = 0.0150, p = 0.0007 and p = 0.0074 respectively) and HV (p = 0.0237, p = 0.0432, p = 0.0005 respectively) groups (Fig. 2C).

Among patients in the Expanded group, 42.8% achieved sustained complete remission after first induction therapy versus 77.4% in the Non-Expanded group. Besides, 28.6% of the Expanded group relapsed after first complete remission versus 12.9% in the Non-Expanded group. Importantly, patients from the Expanded group had significantly poorer overall survival (HR[CI95] = 3.3[0.75–14.7], p = 0.0251) and relapse-free survival (HR[CI95] = 13.1[1.9–87.5], p = 0.0079) after 36 months follow-up (Fig. 2D). Results from the validation cohort were in line with our previous study(8) and demonstrated that CD56^neg CD16⁺ NK cells expansion at diagnosis results in adverse clinical outcome in AML patients.

CD56 ^neg CD16⁺ NK cells display a unique transcriptomic profile of mature circulating NK cells and show altered expression of transcripts involved in NK cell function.

Considering the clinical implications of CD56^neg CD16⁺ NK cells expansion in AML, we aimed to further characterize this subset. To this end, we performed bulk RNA-seq profiling of CD56^neg CD16⁺, CD56^dim CD16⁺, CD56^dim CD16⁻ and CD56^bright NK cells isolated from peripheral blood in N = 5 HV and N = 4 AML patients at diagnosis. NK cell subsets from HV and AML patients were respectively pooled prior to analysis. The purity of CD56^neg CD16⁺ NK cells in our dataset was assessed based on cell type-specific transcriptomic signatures for non-classical monocytes, myeloid dendritic, B and T cells from the human protein atlas(18) (https://www.proteinatlas.org/). Sorted CD56^neg CD16⁺ NK cells expressed some non-classical monocytes-related transcripts, although very slightly, and do not overexpressed transcripts related to myeloid dendritic, B and T cells. Therefore, we concluded that our samples were not contaminated by other cell types (Supplementary Fig. 4). CD56^neg CD16⁺ NK cells from HV and AML patients expressed a unique transcriptomic profile with 177 down-regulated and 161 up-regulated transcripts in AML samples and 87 down-regulated and 252 up-regulated transcripts in HV samples (Supplementary Tables 4 and 5). As expected, NCAM1, encoding for CD56, was down-regulated in CD56^neg CD16⁺ NK cells from both AML patients and HV.

To assess the impact of AML on CD56^neg CD16⁺ NK cells transcriptomic profile, we focused on differentially expressed transcripts unique to the AML group. Venn diagrams identified 120 down-regulated and 53 up-regulated AML-specific transcripts (Fig. 3A, top panel). We first investigated the maturation status of CD56^neg CD16⁺ NK cells from AML patients. This subset down-regulated MYC, involved in NK cell development(19) but up-regulated NFIL3, an essential nuclear factor for NK cell maturation(20). During maturation, NK cells lose CD62L (SELL) and NKG2A (KLRC1) expression and acquire KIRs(21). Consistently, CD56^neg CD16⁺ NK cells down-regulated SELL and KLRC1 and up-regulated KIR2DL1, KIR3DL1, KIR3DL2 and KIR2DS4. Besides, DLL1, which inhibition is involved in Notch-mediated KIRs expression(22), was also down-regulated. Therefore, CD56^neg CD16⁺ NK cells from AML patients seemed to have reached later maturation stage. CD56^neg CD16⁺ NK cells expansion could be a consequence of impaired chemokine or cytokine signalling. We observed that CD56^neg CD16⁺ NK cells from AML patients down-regulated CCR1 and CCR5, involved in NK cell recruitment in inflamed tissues, and up-regulated CCL3 and CCL4, known as CCR5 ligands. Notably, THBS1, shown to take part in late NK cell expansion upon TGF-β stimulation(23), was up-regulated along with TNFRSF1B, involved in TNFα-mediated NK cell proliferation. Besides, KLF4, playing a role in murine NK cell survival and maintenance(24), was also up-regulated. Given the adverse clinical outcome of CD56^neg CD16⁺ NK cells expansion in AML, we also explored relevant transcripts related to NK cell cytotoxicity. Among them, TLR2 and GZMH were up-regulated. However, we also found transcripts that could inhibit NK cell-mediated cytotoxicity. CD56^neg CD16⁺ NK cells down-regulated INF2 described as a key factor of T cells immune synapse(25). In addition, ATF3, a negative regulator of IFN-γ genes in NK cells(26), was up-regulated along with CST7. This molecule is known to be an inhibitor of cathepsins C and H and could promote NK cell split anergy(27) (Fig. 3A, bottom panel). Enrichment analysis on AML-specific transcripts mostly unveiled that up-regulated transcripts were enriched in cytokine-related transcripts (TNF, IL-12, and type II IFN) whereas down-regulated transcripts were enriched in cell activation-related transcripts (Fig. 3B).

Then, we sought to identify a specific transcriptomic signature for CD56^neg CD16⁺ NK cells based on the overlapping transcripts between AML and HV samples. A set of 57 down-regulated and 108 up-regulated transcripts were found (Fig. 3C, top panel). We observed the differential expression of cytotoxicity-related transcripts. CD56^neg CD16⁺ NK cells up-regulated LYZ, the activating receptor CD86(28) and the inhibitory receptor CSF3R(29). Besides, transcripts from the LILR family (LILRA1, LILRA2 and LILRB2) were also up-regulated. On the other hand, CD56^neg CD16⁺ NK cells differentially expressed transcripts that could hamper NK cell-mediated killing. Namely, TMEM163 implied in NK cell degranulation(30) was down-regulated whereas SIGLEC10, which is thought to decrease NK cell cytotoxicity in hepatocellular carcinoma(31), was up-regulated. Interestingly, complement transcripts C3, C3AR1, which was shown to inhibit NK cell cytotoxicity, as well as C5AR1 with immunoregulatory properties in mice(32) were up-regulated. Next, we investigated transcripts involved in NK cell maturation. CD56^neg CD16⁺ NK cells down-regulated transcripts expressed in immature CD56^bright NK cells such as GZMK(33) or KIT(34). They also down-regulated RUNX2, repressed during canonical maturation, along with its target genes ZEB1, BACH2 and PRDM8(35). On the other hand, ZNF683, known to regulated NK cell development(36) but also involved in NK cell exhaustion in multiple myeloma(37), was up-regulated. Then, we assessed the homing abilities of CD56^neg CD16⁺ NK cells. ITGA1, a specific marker of NK cell tissue residency(38), was down-regulated along with CCR7(39) and CXCR3(40). Besides, CXCR2 and its ligands CXCL8 and CXCL16(41), IL3RA and CD4, increasing cytokine production and cell migration in NK cells(42), were up-regulated (Fig. 3C, bottom panel). Furthermore, cell activation, cytokine pathways, proliferation and endocytosis were overrepresented among over-expressed transcripts while cell activation, chemotaxis, actin organization and Wnt signalling were overrepresented among under-expressed transcripts as revealed by gene set enrichment analysis from subset-specific transcripts (Fig. 3D). Together these results strongly suggested that CD56^neg CD16⁺ NK cells represent a unique subset of mature NK cells with functional capacities relying on different cytotoxic pathways than conventional NK cells. Notably, some transcripts were related to impaired cytotoxicity or exhaustion, which could partly explain the adverse outcome in AML patients with CD56^neg CD16⁺ NK cells.

Altered phenotype of CD56 ^neg CD16⁺ NK cells in AML patients with CD56^neg CD16⁺ NK cells expansion.

After RNA-seq profiling, we attempted to confirm our findings with spectral flow cytometry and investigated whether CD56^neg CD16⁺ NK cells had a distinct protein expression pattern than conventional NK cells in HV, Non-Expanded and Expanded groups.

CD56^neg CD16⁺ NK cells from all three groups had significantly decreased expression of NK cell triggering receptors NKp30 and NKp46 compared to CD56^bright NK cells. However, CD56^neg CD16⁺ NK cells from the Expanded group expressed similar levels of NKp30 and NKp46 compared to CD56^dim NK cells. CD56^neg CD16⁺ NK cells from the HV group had decreased NKp46 expression compared to CD56^dim CD16⁻ NK cells whereas CD56^neg CD16⁺ NK cells from the Non-Expanded group had decreased NKp46 expression compared to CD56^dim CD16⁺ NK cells.

Regarding direct cytotoxicity, CD56^neg CD16⁺ NK cells from the HV and Expanded groups significantly expressed higher levels of granzyme B and perforin compared to CD56^dim CD16⁻ NK cells. CD56^neg CD16⁺ NK cells from the HV and Non-Expanded groups significantly expressed higher levels of granzyme B and perforin compared to CD56^bright NK cells. CD56^neg CD16⁺ NK cells from the Non-Expanded group had decreased expression of granzyme B and perforin compared to CD56^dim CD16⁺ NK cells. On the other hand, CD56^neg CD16⁺ NK cells from the Expanded group had similar expression of granzyme B and perforin than CD56^dim CD16⁺ NK cells. Thus, our results are in line with both our transcriptomic and in vitro data : CD56^neg CD16⁺ NK cells displayed a cytotoxic profile in AML. On the other hand, ours results confirmed that CD56^neg CD16⁺ NK cells had reached a mature stage as CD56^neg CD16⁺ NK cells from all three groups expressed significantly lower levels of NKG2A and higher levels of CD158a/b than CD56^bright NK cells. Moreover, CD56^neg CD16⁺ NK cells from the HV and Non-Expanded groups had decreased expression of NKG2A compared to CD56^dim CD16⁻ NK cells (Fig. 4A).

Next, we investigated whether CD56^neg CD16⁺ NK cells from the Expanded group displayed an altered phenotype compared to the HV and Non-Expanded groups. We performed UMAP analysis of total NK cells and automatic clustering using FlowSOM, resulting in 12 clusters identified by the co-expression of 27 markers (Fig. 4B, left panel, Supplementary Fig. 5). We plotted CD56^neg CD16⁺ NK cells density for each group and observed changes in CD56^neg CD16⁺ NK cells from the Expanded group compared to the HV and Non-Expanded groups (Fig. 4B, right panel). More CD56^neg CD16⁺ NK cells from the Expanded group were found in mature cytotoxic clusters expressing granzyme B and perforin compared to CD56^neg CD16⁺ NK cells from the HV and Non-Expanded groups. Indeed, cluster 2 (TIM-3⁺ CD158a/b⁺ Siglec-7⁺), cluster 3 (TIM-3⁺ CD158a/b⁺ NKG2D⁺ DNAM-1⁺), cluster 4 (TIM-3⁺ CD158a/b⁻ CD57⁺) and cluster 12 containing adaptative NK cells (TIM-3⁻TIGIT⁺ CD158a/b⁺ CD57⁺ NKG2C⁺) (Fig. 4C) were more abundant in the Expanded group than in the HV and Non-Expanded group (Fig. 4D). Besides, CD56^neg CD16⁺ NK cells found in cluster 6 (TIGIT⁺ CD158a/b⁺ Siglec-7⁺) and in cluster 7 (TIM-3⁺ CD158a/b⁻) (Fig. 4C) were significantly more abundant in the Expanded group compared to the Non-Expanded group (Fig. 4D). Finally, cluster 9 (TIM-3⁺ granzyme B⁺ perforin⁺) was decreased in the Expanded group compared to the HV group (Fig. 4D). Interestingly, the expression of TIM-3 and Siglec-7 are expressed by mature NK cells, which confirms that CD56^neg CD16⁺ NK cells are mature NK cells. Besides, TIM-3 and Siglec-7 expression on NK cells were shown to mark fully functional NK cells but could also inhibit NK cell activation when engaged with their cognate ligand(43, 44). Together, these results showed an altered phenotype of CD56^neg CD16⁺ NK cells from the Expanded group, with increased co-expression of proteins involved in NK cell cytotoxicity as well as inhibitory receptors such as CD158a/b, TIM-3, and Siglec-7.

In this study, our aim was to further characterize CD56^neg CD16⁺ NK cells, a subset that expands in AML. We confirmed that CD56^neg CD16⁺ NK cells represent a unique NK cell subset. We used Cytosplore^+ HSNE, an automated clustering approach, on mass cytometry data and observed that CD56^neg CD16⁺ NK cells located in NK cells cluster. These results suggested a strong similarity between proteomic phenotype of CD56^neg CD16⁺ and conventional NK cells, in line with previous publications(2, 45). Besides, CD56^neg CD16⁺ NK cells co-expressed high levels of Eomes and T-bet, distinguishing them from ILCs(46). CD56^neg CD16⁺ NK cells seem to be circulating NK cells as they down-regulated transcripts involved in homing or tissue residency such as CCR7 and ITGA1. Interestingly, CD56^neg CD16⁺ NK cells from AML patients up-regulated THBS1 and TNFRSF1B involved in TGF-β and TNFα-mediated expansion, respectively. Those two cytokines were reported to be increased in AML patients compared to

volunteers(47, 48).

We further unveiled that CD56^neg CD16⁺ NK cells downregulate NCAM1, encoding for CD56, compared to conventional NK cells, suggesting that the loss of CD56 is an intrinsic mechanism. The loss of CD56 was reversible as CD56^neg CD16⁺ NK cells could regain CD56 expression when cultivated with IL-2, IL-15 and IL-21. Importantly, CD56^neg CD16⁺ NK cells displayed unaltered NK cell function in vitro once they regained CD56, known to play a critical role in NK cell cytotoxicity(49).

On the other hand, CD56^negCD16⁺ NK cells expansion has been documented in chronic diseases. In a validation cohort, we confirmed that newly-diagnosed AML patients with CD56^neg CD16⁺ NK cells expansion had significantly poorer clinical outcomes than others. This affords the possibility for adoptive NK cell therapy or the use of NK-cell engagers on CD56^neg CD16⁺ NK cells from AML patients.

Bulk RNA-seq and spectral flow cytometry results showed that CD56^neg CD16⁺ NK cells poorly expressed NKG2A and up-regulated KIRs. Therefore, this subset has reached late maturation stage. Importantly, we observed that CD56^neg CD16⁺ NK cells overexpressed ZNF683. This transcription factor was recently identified as a driver of NK cell exhaustion in multiple myeloma(37), another haematological malignancy. The authors performed pseudotime analysis on ZNF683⁺ NK cells and underlined 5 transcripts indicative of exhaustion state. Among them, 4 were found up-regulated in CD56^neg CD16⁺ NK cells of AML patients, namely ZNF683, LAG3, FGL2 and CST7, involved in NK cell split anergy(27). Furthermore, CD56^neg CD16⁺ NK cells from patients with expansion displayed increased frequency of TIM-3 and Siglec-7 compared to CD56^neg CD16⁺ NK cells from HV and patients without expansion. These two markers are expressed by terminally differentiated NK cells with high cytotoxic potential but can lead to NK cell inhibition depending on their bound ligand. Taken together, our results suggested that AML microenvironment could decrease CD56^neg CD16⁺ NK cells cytotoxicity, which may partly explain the negative impact of CD56^neg CD16⁺ NK cells expansion in AML on survival.

There were some limitations in the study. RNA-seq analysis was performed on pooled samples and without biological replicates. Therefore, we were likely to find high false positive rates in our differential gene expression analysis. Nevertheless, our results corroborated previous studies(2, 5). Besides, CMV status from AML patients were not available and we cannot exclude the possibility that CD56^neg CD16⁺ NK cells expansion is driven by CMV infection(50). However, adaptative CD57⁺ NKG2C⁺ NK cells from AML patients with CD56^neg CD16⁺ NK cells expansion were not significantly increased compared to the AML patients without expansion (data not shown).

To conclude, we demonstrated that CD56^neg CD16⁺ NK cells represent a distinct NK cell subset that can recover CD56 expression in vitro and display NK cell functions, which could be leveraged to counteract the adverse clinical outcome of AML patients with CD56^neg CD16⁺ NK cells expansion.

Acknowledgements

We would like to thank all patients that participated in this study. We thank the Gladstone Bioinformatics Core for their interactive enrichment analysis tools, the CRCM cytometry facility core as well as the Paoli-Calmettes Tumor Bank. This article is supported by grants from Assistance Publique-Hôpitaux de Marseille (AP-HM) and by the Association Laurette Fugain (grant ALF2020/05 for D.O.).The team “Immunity and Cancer” was labelled Equipe Fondation pour la Recherche Médicale (FRM) (DEQ20180339209 for D.O). Figures were generated with BioRender.com.

Authorship Contributions

J.W. designed experiments, performed experiments, analysed data, and wrote the manuscript. F.O., M.R., C.D., A.B.A., M.S.R., P.L. performed experiments. M.R., C.D., A.B.A., M.S.R., L.G., M.A.H., N.D. and N.V. reviewed the manuscript. D.O. and A.S.C. designed research and reviewed the manuscript.

Conflict of Interest Disclosures

D.O. is a cofounder and shareholder of ImCheck Therapeutics, Alderaan Biotechnology and Emergence Therapeutics and has research funds from ImCheck Therapeutics, Alderaan Biotechnology, Cellectis and Emergence Therapeutics. The other authors declare no conflicts of interest.

Data sharing statement

Data are available upon reasonable request from the corresponding authors.

Boudreau JE, Hsu KC. Natural Killer Cell Education and the Response to Infection and Cancer Therapy: Stay Tuned. Trends in Immunology. 2018 Mar 1;39(3):222–39.
Forconi CS, Oduor CI, Oluoch PO, Ong’echa JM, Münz C, Bailey JA, et al. A New Hope for CD56negCD16pos NK Cells as Unconventional Cytotoxic Mediators: An Adaptation to Chronic Diseases. Front Cell Infect Microbiol. 2020;10:162.
Hu PF, Hultin LE, Hultin P, Hausner MA, Hirji K, Jewett A, et al. Natural killer cell immunodeficiency in HIV disease is manifest by profoundly decreased numbers of CD16+CD56+ cells and expansion of a population of CD16dimCD56- cells with low lytic activity. J Acquir Immune Defic Syndr Hum Retrovirol. 1995 Nov 1;10(3):331–40.
Mavilio D, Lombardo G, Benjamin J, Kim D, Follman D, Marcenaro E, et al. Characterization of CD56–/CD16+ natural killer (NK) cells: A highly dysfunctional NK subset expanded in HIV-infected viremic individuals. Proceedings of the National Academy of Sciences. 2005 Feb 22;102(8):2886–91.
Cao WJ, Zhang XC, Wan LY, Li QY, Mu XY, Guo AL, et al. Immune Dysfunctions of CD56neg NK Cells Are Associated With HIV-1 Disease Progression. Front Immunol. 2021;12:811091.
Gonzalez VD, Falconer K, Björkström NK, Blom KG, Weiland O, Ljunggren HG, et al. Expansion of functionally skewed CD56-negative NK cells in chronic hepatitis C virus infection: correlation with outcome of pegylated IFN-alpha and ribavirin treatment. J Immunol. 2009 Nov 15;183(10):6612–8.
Ty M, Sun S, Callaway PC, Rek J, Press KD, van der Ploeg K, et al. Malaria-driven expansion of adaptive-like functional CD56-negative NK cells correlates with clinical immunity to malaria. Sci Transl Med. 2023 Jan 25;15(680):eadd9012.
Chretien AS, Devillier R, Granjeaud S, Cordier C, Demerle C, Salem N, et al. High-dimensional mass cytometry analysis of NK cell alterations in AML identifies a subgroup with adverse clinical outcome. Proceedings of the National Academy of Sciences. 2021 Jun;118(22):e2020459118.
D’Silva SZ, Singh M, Pinto AS. NK cell defects: implication in acute myeloid leukemia. Frontiers in Immunology [Internet]. 2023 [cited 2023 Oct 2];14. Available from: https://www.frontiersin.org/articles/10.3389/fimmu.2023.1112059
Lion E, Willemen Y, Berneman ZN, Van Tendeloo VFI, Smits ELJ. Natural killer cell immune escape in acute myeloid leukemia. Leukemia. 2012 Sep;26(9):2019–26.
Kantarjian H, Kadia T, DiNardo C, Daver N, Borthakur G, Jabbour E, et al. Acute myeloid leukemia: current progress and future directions. Blood Cancer J. 2021 Feb 22;11(2):1–25.
Xu J, Niu T. Natural killer cell-based immunotherapy for acute myeloid leukemia. Journal of Hematology & Oncology. 2020 Dec 7;13(1):167.
Gunesch JT, Dixon AL, Ebrahim TA, Berrien-Elliott MM, Tatineni S, Kumar T, et al. CD56 regulates human NK cell cytotoxicity through Pyk2. Elife. 2020 Jun 8;9:e57346.
Taouk G, Hussein O, Zekak M, Abouelghar A, Al-Sarraj Y, Abdelalim EM, et al. CD56 expression in breast cancer induces sensitivity to natural killer-mediated cytotoxicity by enhancing the formation of cytotoxic immunological synapse. Sci Rep. 2019 Jun 19;9(1):8756.
Gauthier L, Virone-Oddos A, Beninga J, Rossi B, Nicolazzi C, Amara C, et al. Control of acute myeloid leukemia by a trifunctional NKp46-CD16a-NK cell engager targeting CD123. Nat Biotechnol. 2023 Sep;41(9):1296–306.
van Unen V, Höllt T, Pezzotti N, Li N, Reinders MJT, Eisemann E, et al. Visual analysis of mass cytometry data by hierarchical stochastic neighbour embedding reveals rare cell types. Nat Commun. 2017 Nov 23;8(1):1740.
Wong P, Foltz JA, Chang L, Neal CC, Yao T, Cubitt CC, et al. T-BET and EOMES sustain mature human NK cell identity and antitumor function. J Clin Invest [Internet]. 2023 Jul 3 [cited 2024 May 7];133(13). Available from: https://www.jci.org/articles/view/162530
Uhlen M, Karlsson MJ, Zhong W, Tebani A, Pou C, Mikes J, et al. A genome-wide transcriptomic analysis of protein-coding genes in human blood cells. Science. 2019 Dec 20;366(6472):eaax9198.
Khameneh HJ, Fonta N, Zenobi A, Niogret C, Ventura P, Guerra C, et al. Myc controls NK cell development, IL-15-driven expansion, and translational machinery. Life Sci Alliance. 2023 Apr 27;6(7):e202302069.
Kamizono S, Duncan GS, Seidel MG, Morimoto A, Hamada K, Grosveld G, et al. Nfil3/E4bp4 is required for the development and maturation of NK cells in vivo. Journal of Experimental Medicine. 2009 Dec 7;206(13):2977–86.
Björkström NK, Riese P, Heuts F, Andersson S, Fauriat C, Ivarsson MA, et al. Expression patterns of NKG2A, KIR, and CD57 define a process of CD56dim NK-cell differentiation uncoupled from NK-cell education. Blood. 2010 Nov 11;116(19):3853–64.
Felices M, Ankarlo DEM, Lenvik TR, Nelson HH, Blazar BR, Verneris MR, et al. Notch signaling at later stages of Natural Killer cell development enhances KIR expression and functional maturation. J Immunol. 2014 Oct 1;193(7):3344–54.
Pierson BA, Gupta K, Hu WS, Miller JS. Human Natural Killer Cell Expansion Is Regulated by Thrombospondin-Mediated Activation of Transforming Growth Factor-β1 and Independent Accessory Cell-Derived Contact and Soluble Factors. Blood. 1996 Jan 1;87(1):180–9.
Shik Park C, Lee PH, Yamada T, Burns A, Shen Y, Puppi M, et al. Krüppel-like factor 4 (KLF4) promotes the survival of natural killer cells and maintains the number of conventional dendritic cells in the spleen. J Leukoc Biol. 2012 May;91(5):739–50.
Andrés-Delgado L, Antón OM, Bartolini F, Ruiz-Sáenz A, Correas I, Gundersen GG, et al. INF2 promotes the formation of detyrosinated microtubules necessary for centrosome reorientation in T cells. J Cell Biol. 2012 Sep 17;198(6):1025–37.
Rosenberger CM, Clark AE, Treuting PM, Johnson CD, Aderem A. ATF3 regulates MCMV infection in mice by modulating IFN-γ expression in natural killer cells. Proc Natl Acad Sci U S A. 2008 Feb 19;105(7):2544–9.
Perišić Nanut M, Sabotič J, Švajger U, Jewett A, Kos J. Cystatin F Affects Natural Killer Cell Cytotoxicity. Front Immunol. 2017 Nov 13;8:1459.
Peng Y, Luo G, Zhou J, Wang X, Hu J, Cui Y, et al. CD86 Is an Activation Receptor for NK Cell Cytotoxicity against Tumor Cells. PLoS One. 2013 Dec 11;8(12):e83913.
Cao X, Han TT, Zhao M, Juan X, Chang YJ, Huang XJ, et al. G-CSFR Is a Checkpoint of Natural Killer Cells Against Tumor. Blood. 2023 Nov 28;142(Supplement 1):3923.
Gilchrist JJ, Makino S, Naranbhai V, Sharma PK, Koturan S, Tong O, et al. Natural Killer cells demonstrate distinct eQTL and transcriptome-wide disease associations, highlighting their role in autoimmunity. Nat Commun. 2022 Jul 14;13(1):4073.
Zhang P, Lu X, Tao K, Shi L, Li W, Wang G, et al. Siglec-10 is associated with survival and natural killer cell dysfunction in hepatocellular carcinoma. Journal of Surgical Research. 2015 Mar 1;194(1):107–13.
Min X, Liu C, Wei Y, Wang N, Yuan G, Liu D, et al. Expression and regulation of complement receptors by human natural killer cells. Immunobiology. 2014 Sep 1;219(9):671–9.
Cheng S, Li Z, Gao R, Xing B, Gao Y, Yang Y, et al. A pan-cancer single-cell transcriptional atlas of tumor infiltrating myeloid cells. Cell. 2021 Feb 4;184(3):792-809.e23.
Matos ME, Schnier GS, Beecher MS, Ashman LK, William DE, Caligiuri MA. Expression of a functional c-kit receptor on a subset of natural killer cells. J Exp Med. 1993 Sep 1;178(3):1079–84.
Holmes TD, Pandey RV, Helm EY, Schlums H, Han H, Campbell TM, et al. The transcription factor Bcl11b promotes both canonical and adaptive NK cell differentiation. Sci Immunol. 2021 Mar 12;6(57):eabc9801.
Post M, Cuapio A, Osl M, Lehmann D, Resch U, Davies DM, et al. The Transcription Factor ZNF683/HOBIT Regulates Human NK-Cell Development. Front Immunol. 2017;8:535.
Li X, Chen M, Wan Y, Zhong L, Han X, Chen X, et al. Single-cell transcriptome profiling reveals the key role of ZNF683 in natural killer cell exhaustion in multiple myeloma. Clinical and Translational Medicine. 2022;12(10):e1065.
Hegewisch-Solloa E, Seo S, Mundy-Bosse BL, Mishra A, Waldman ErikH, Maurrasse S, et al. Differential integrin adhesome expression defines human natural killer cell residency and developmental stage. J Immunol. 2021 Aug 1;207(3):950–65.
Pesce S, Moretta L, Moretta A, Marcenaro E. Human NK Cell Subsets Redistribution in Pathological Conditions: A Role for CCR7 Receptor. Front Immunol. 2016 Oct 7;7:414.
Ali A, Canaday LM, Feldman HA, Cevik H, Moran MT, Rajaram S, et al. Natural killer cell immunosuppressive function requires CXCR3-dependent redistribution within lymphoid tissues. J Clin Invest. 131(18):e146686.
Susek KH, Karvouni M, Alici E, Lundqvist A. The Role of CXC Chemokine Receptors 1–4 on Immune Cells in the Tumor Microenvironment. Front Immunol [Internet]. 2018 Sep 25 [cited 2024 May 15];9. Available from: https://www.frontiersin.org/journals/immunology/articles/10.3389/fimmu.2018.02159/full
Bernstein HB, Plasterer MC, Schiff SE, Kitchen CMR, Kitchen S, Zack JA. CD4 expression on activated NK cells: ligation of CD4 induces cytokine expression and cell migration. J Immunol. 2006 Sep 15;177(6):3669–76.
Ndhlovu LC, Lopez-Vergès S, Barbour JD, Jones RB, Jha AR, Long BR, et al. Tim-3 marks human natural killer cell maturation and suppresses cell-mediated cytotoxicity. Blood. 2012 Apr 19;119(16):3734–43.
Shao JY, Yin WW, Zhang QF, Liu Q, Peng ML, Hu HD, et al. Siglec-7 Defines a Highly Functional Natural Killer Cell Subset and Inhibits Cell-Mediated Activities. Scandinavian Journal of Immunology. 2016;84(3):182–90.
Voigt J, Malone DFG, Dias J, Leeansyah E, Björkström NK, Ljunggren HG, et al. Proteome analysis of human CD56neg NK cells reveals a homogeneous phenotype surprisingly similar to CD56dim NK cells. European Journal of Immunology. 2018;48(9):1456–69.
Zhang J, Marotel M, Fauteux-Daniel S, Mathieu AL, Viel S, Marçais A, et al. T-bet and Eomes govern differentiation and function of mouse and human NK cells and ILC1. European Journal of Immunology. 2018;48(5):738–50.
Wang C zhu, Zhang Z qi, Zhang Y, Zheng L feng, Liu Y, Yan A ting, et al. Comprehensive characterization of TGFB1 across hematological malignancies. Sci Rep. 2023 Nov 4;13(1):19107.
Sanchez-Correa B, Bergua JM, Campos C, Gayoso I, Arcos MJ, Bañas H, et al. Cytokine profiles in acute myeloid leukemia patients at diagnosis: Survival is inversely correlated with IL-6 and directly correlated with IL-10 levels. Cytokine. 2013 Mar 1;61(3):885–91.
Gunesch JT, Dixon AL, Ebrahim TA, Berrien-Elliott MM, Tatineni S, Kumar T, et al. CD56 regulates human NK cell cytotoxicity through Pyk2. Elife. 2020 Jun 8;9:e57346.
Müller-Durovic B, Grählert J, Devine OP, Akbar AN, Hess C. CD56-negative NK cells with impaired effector function expand in CMV and EBV co-infected healthy donors with age. Aging (Albany NY). 2019 Jan 27;11(2):724–40.

Table 1. Patients’ characteristics.

Characteristics	AML		Non-Expanded group	Expanded group
Patients, No. (%)		38 (100)	31 (81.6)		7 (18.4)
Age at diagnosis, mean (SD)		49 (15.6)	49.3 (16.1)		47.7 (13.8)
Sex, No. (%)
Female		23 (60.5)	20 (64.5)		3 (42.9)
Male		15 (39.5)	11 (35.5)		4 (57.1)
ELN, No. (%)
Favorable		14 (36.8)	13 (41.9)		1 (14.3)
Intermediate		10 (26.3)	8 (25.8)		2 (28.6)
Adverse		6 (15.8)	3 (9.7)		3 (42.8)
NA		8 (21.1)	7 (22.6)		1 (14.3)
Treatment, No. (%)
Chemotherapy		38 (100)	31 (100)		7 (100)
Chemotherapy + Allo-SCT		16 (42.1)	11 (35.5)		5 (71.4)
Treatment response, No. (%)
CR		27 (71.0)	24 (77.4)		3 (42.8)
Relapse		6 (15.8)	4 (12.9)		2 (28.6)
Refractory		5 (13.2)	3 (9.7)		2 (28.6)
Blasts (bone marrow) at diagnosis, mean (SD)		58.6 (26.2)	57.1 (26.8)		65.3 (24.1)
Blasts (blood) at diagnosis, mean (SD)		30.7 (31.0)	32.4 (31.6)		23.1 (29.1)

Yes there is potential conflict of interest.

Download PDF

Version 1

posted

You are reading this latest preprint version

CD56^negCD16⁺ cells represent a distinct mature NK cell subset with altered phenotype and are associated with adverse clinical outcome upon expansion in AML.

Status:

Version 1

Abstract

Figures

Introduction

Materials and Methods

Results

Discussion

Declarations

References

Table 1

Additional Declarations

Supplementary Files

Status:

Version 1