Intracellular IL-23R is necessary for mitotic spindle formation and viability in AML

Leukemia ARTICLE www.nature.com/leu OPEN Intracellular IL-23R is necessary for mitotic spindle formation and viability in AML Nathan Duong 1,2, Dilshad H. Khan1, Geethu E. Thomas1, Yue Feng1,2, Rose Hurren1, Jong Bok Lee3, Jonathan St-Germain1, Lily Drimmer1, Yongran Yan 1, Lan Xin Zhang1,2, Karen Kai-Lin Fang1, Dakai Ling1, Mary L. Ma1, Neil MacLean1, Marcela Gronda1, Vincent Rondeau1, Brandon D. Brown4, Laura Matellán 5, Courtney L. Jones1, Hong Chang 6, Andrea Arruda1, Stephanie Xie 1, Laurence Pelletier5, Mark D. Minden1,2, Li Zhang3, Steven M. Kornblau 4, Brian Raught1,2, Kevin Jacobs7, Max G. Jacobs7, ✉ Daniel Goede7, Vito Spadavecchio7 and Aaron D. Schimmer 1,2 1234567890();,: © The Author(s) 2026 Interleukin-23 receptor (IL-23R) is a cell surface cytokine receptor classically expressed on T cells, where it regulates T cell activation. Here, we discovered a novel intracellular localization and function for IL-23R in Acute Myeloid Leukemia (AML). Compared to normal hematopoietic cells, IL-23R was increased in primary AML samples. IL-23R was predominantly localized intracellularly in AML cells. BioID mass spectrometry identified mitotic spindle proteins as top interactors with IL-23R. We confirmed interaction between endogenous IL-23R and the mitotic spindle in AML cells and primary AML samples, and this interaction was mediated by IL-23R’s (S/ T)x(I/L)P motif. Genetic depletion of IL-23R disrupted mitotic spindle formation and reduced proliferation and stem cell/progenitor function of AML cell lines and primary AML samples. In contrast, depletion of IL-23R spared normal hematopoietic cells and progenitors. Thus, we discovered a novel intracellular function for IL-23R where this receptor regulates mitotic spindle formation and the growth of AML cells. Leukemia; https://doi.org/10.1038/s41375-026-02949-8 INTRODUCTION IL-23R is a heterodimeric receptor composed of the IL-23R subunit in complex with IL-12Rβ1 [1]. IL-23R is classically found on the surface of T cells, where it is activated by its ligand cytokine, IL-23 [2], a pro-inflammatory cytokine secreted by activated macrophages and dendritic cells located in the skin, intestinal mucosa, and lung in response to environmental stimuli [3]. In one proposed model of cytokine signaling, the p19 subunit of IL-23 binds the IL-23R subunit [4]. Upon binding, IL-23 acts as a bridge and recruits IL-12Rβ1 to the IL-12p40 subunit of IL-23 to form the full IL-23R receptor complex [4]. The binding of IL-23 to IL-23R activates the JAK/STAT signaling domain of the receptor, leading to STAT3 phosphorylation and subsequent T cell activation [2]. Specifically, IL-23R activation amplifies the proliferation of CD4+ memory T cells and the differentiation of naïve T cells into Th17 cells [1, 2]. Aberrant IL-23R signaling through excessive IL-23 production leads to increased generation of Th17 cells with a resultant increase in IL-17 and other inflammatory cytokines [2]. Through this mechanism, increased IL-23 signaling is linked to autoimmune disorders such as psoriasis and inflammatory bowel disease, where IL-23 inhibitors are in clinical use for these diseases [5–7]. In this study, we discovered a novel and unexpected function for IL-23R in Acute Myeloid Leukemia (AML). AML is an aggressive hematologic malignancy characterized by the uncontrollable proliferation of malignant myeloid progenitors. We discovered that IL-23R was located intracellularly in AML cells and stem cells. Intracellularly, IL-23R modulated cell proliferation and viability by interacting with and regulating the formation of the mitotic spindle and centrosomes via its (S/T)x(I/L)P motif in its cytoplasmic domain. Genetic depletion of IL-23R led to mitotic defects, primarily unaligned chromosomes, before subsequent cell death in AML. Thus, we describe a non-canonical function for IL-23R and a novel dependency on this cytokine receptor in AML. METHODS Immunoblotting Whole cell lysates, nuclear, or cytoplasmic lysates from cell lines or primary patient samples were lysed using radioimmunoprecipitation assay (RIPA) buffer before protein quantification using the Bradford assay (Bio-Rad). Equal amounts of protein were run on 10–12% SDS-PAGE gels before transfer to polyvinylidene difluoride membranes. Membranes were then blocked in 5% skim milk in Tris-buffered saline with Tween-20 (TBST) for 1 h. Blocked membranes were then incubated overnight at 4 °C with primary antibody dissolved in 5% skim milk in TBST. Membranes were then washed three times with TBST and incubated with respective secondary antibodies for 1 h at room temperature. Densitometry was performed on Bio-Rad’s Image Lab (v6.0.1 build 34). Antibody information is described in the supplementary methods. 1 Princess Margaret Cancer Centre, University Health Network, Toronto, ON, Canada. 2Department of Medical Biophysics, University of Toronto, Toronto, ON, Canada. 3Toronto General Hospital Research Institute, University Health Network, Toronto, ON, Canada. 4MD Anderson Cancer Centre, Houston, TX, USA. 5Lunenfeld-Tanenbaum Research Institute, Mount Sinai Hospital, Toronto, ON, Canada. 6Laboratory Medicine Program, Department of Laboratory Hematology, University Health Network, Toronto, ON, Canada. 7Interlinked Therapeutics, Portland, OR, USA. ✉email: Received: 10 December 2025 Revised: 11 March 2026 Accepted: 27 March 2026 N. Duong et al. 2 Reverse-phase protein array Samples were collected from 810 patients with AML. The MD Anderson Cancer Center Institutional Review Board approved the collection protocol, research usage profile and clinical protocols that these patients were treated with. Protein expression levels of samples from patients with AML were determined using RPPA analysis. Methods and antibody validation techniques have been fully described previously [8–10]. Leukemia N. Duong et al. Fig. 1 IL-23R expression and localization in AML and stem cells. a Most significant Molecular Signatures Database (MSigDB) Hallmark ontology enrichments for comparison of AML samples to control. b Outdegree of each upregulated ontology in a directed acyclic graph representation of AML signaling. c Enrichment scores for the regulation of the mitotic spindle ontology in rank order for 18,795 proteincoding genes from an in-silico screen. d Densitometry of IL-23R expression between CD34 + PBSC samples (n = 3), bulk PBSC samples (n = 5), and primary AML samples (n = 20) relative to sample OCI 130853. Primary AML vs CD34 + PBSC (*p = 0.0369). Primary AML vs bulk PBSC (*p = 0.0373). Statistical analyses were performed using a two-tailed unpaired Student’s t test (t = 2.229, df=21). Data represent the mean ± SD. e Geometric mean fluorescence of IL-23R in intact leukemia cell lines (HL60, AML2, NB4, TEX, U937, K562, 8227, and CD4 and CD8 double negative T (DNT) cells) (n = 3) above unstained controls via flow cytometry. Values in brackets represent the geometric mean of unstained controls. Data represent the mean ± SD. f (...truncated)