AUTHORS
Erin Jeremy, MS, Amy Zhang2, Alessandro La Ferlita, PhD3, Michael Ruesch4, Noah Koenigs5, Kyoko Yamaguchi, BSc6, Yue-Zhong Wu7, Antonio Perez-Castro, PhD8, Thanh Thanh T. Dinh, PhD9, Antony Shilo10, Matthew Lordo, MD, PhD11, Salma Abdelbaky, PhD12, Patrick Collins, PhD13, Karilyn Larkin, MD14, Aharon G Freud, MD, PhD15, Christopher C Oakes, PhD16, Bethany Mundy-Bosse, PhD17
AFFILIATIONS
1 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
2 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
3 The Ohio State University, Columbus, Ohio, United States
4 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
5 The Ohio State University, Columbus, Ohio, United States
6 The Ohio State University, Columbus, Ohio, United States
7 The Ohio State University, Columbus, Ohio, United States
8 The Ohio State University, Columbus, Ohio, United States
9 The Ohio State University, Columbus, Ohio, United States
10 The Ohio State University, Columbus, Ohio, United States
11 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
12 The Ohio State University, Columbus, Ohio, United States
13 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
14 The Ohio State University, Columbus, Ohio, United States
15 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
16 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
17 The Ohio State University, Columbus, Ohio, United States, The Ohio State University, Columbus, Ohio, United States
ABSTRACT
The natural killer (NK) cell landscape is profoundly altered in acute myeloid leukemia (AML) and is marked by impairments in differentiation and effector function that are associated with adverse clinical outcomes. While it has been well established that NK cells are dysfunctional in AML, the mechanism(s) that underly these defects are poorly understood. To elucidate pathway(s) that are responsible for the defects in NK cell development, differentiation and function, we evaluated epigenetic, transcriptional, and functional changes in NK precursors and mature NK cells from AML patients using scRNAseq, DNA Illumina methylation analyses, CUT&RUN technology, and high-dimensional flow cytometry.
We first evaluated the DNA methylation profile of NK cell populations found in peripheral blood (PB) and discovered profound hypomethylation of AP-1 transcription factor (TF) motifs in NK cells isolated from AML patients as compared to healthy donors (HD; p<0.0001). scRNAseq of NK cells from HD and AML patients further revealed two specific clusters that were expanded in AML (p<0.001) which were driven by upregulation of multiple AP-1 TFs including JUN, JUNB, FOS and FOSB (all genes; p<0.05). These NK cell clusters exhibited increased activation but a decreased cytotoxic transcriptional profile (p<0.05), which we termed Dysfunctional and ACtivated (NKDAC). Enrichment of the NKDAC signature was also confirmed using an independent single-cell dataset (p<0.0001). Sequencing and flow cytometry revealed increased expression of inhibitory receptors, including KIR, TIM-3, TIGIT and CD45RO (p<0.05), consistent with their dysfunctional phenotype. Additional scRNAseq performed on NK cells derived from a murine model of AML that recapitulates the developmental and phenotypical changes seen in patients demonstrated increased AP-1 TF expression in NK precursors (p<0.001) and a subsequent enrichment of our NKDAC signature (p<2e-16).
AP-1 is a transcription factor dimer family involved in dysfunctional immune programs such as chronic activation, inflammaging, and exhaustion and is regulated by MAPK signaling (JNK, p38, MEK/ERK). Given the aberrant AP-1 activity in NK precursors and mature NK cells, we targeted each major MAPK branch with pharmacological inhibitors and discovered that only MEK inhibition restored the development of NK cells generated from CD34+ HPCs (p<0.05) and innate lymphoid cell-restricted precursors (ILCPs; p<0.05). Additionally, we designed a co-culture system utilizing HD NK cells and AML that recapitulates the upregulation of inhibitory markers observed in patients (KIR, TIM-3, CD45RO) and found that MEK inhibition also prevented this dysfunctional phenotype (p<0.01).
Next, we sought to identify which AP-1 TF might drive the NKDAC program. JUNB, a highly dynamic TF associated with immune cell dysfunction, was consistently upregulated in all 3 scRNAseq datasets evaluated; thus, we hypothesized that aberrant JUNB activity could regulate NK cell development and/or NKDAC programming in AML. Using a viral-like particle system to introduce JUNB-directed Cas9 into NK cell precursors to selectively knock out JUNB, we found that JUNB-KO alone restored the loss of NK cells generated from CD34+ HPCs in the setting of AML (p<0.05). CUT&RUN-seq on HD NK cells co-cultured with and without AML demonstrated differential JUNB binding induced by the presence of AML (p<0.05). Specifically, genes showing increased JUNB occupancy were significantly enriched for the NKDAC gene signature (p<0.01), binding to genes such as CD69, PTGER4, and TNFAIP3.
Collectively, we demonstrate that aberrant and persistent MEK/ERK/AP-1 signaling reprograms the fate and function of NK precursors and committed NK cells in AML. Aberrant MEK/ERK activity in AML prevents the development of NK cells from NK precursors and impairs the function of committed NK cells by inducing an NKDAC program which is evident at the epigenetic, transcriptional, and protein level. Finally, we show that disrupting the MEK/ERK/AP-1 signaling axis and specific depletion of JUNB restores NK cell development and prevents NKDAC programming in primary NK cells. This work demonstrates that aberrant MEK/ERK/AP-1 signaling is a novel, targetable pathway in AML that could potentially restore NK cell immune surveillance.
DISCLOSURES
Amy Zhang: No relevant conflicts of interest to declare.