Structure of the interleukin-5 receptor complex exemplifies the organizing principle of common beta cytokine signaling

Abstract

Cytokines regulate immune responses by binding to cell surface receptors, including the common subunit beta (βc), which mediates signaling for GM-CSF, IL-3, and IL-5. Despite known roles in inflammation, the structural basis of IL-5 receptor activation remains unclear. We present the cryo-EM structure of the human IL-5 ternary receptor complex, revealing architectural principles for IL-5, GM-CSF, and IL-3. In mammalian cell culture, single-molecule imaging confirms hexameric IL-5 complex formation on cell surfaces. Engineered chimeric receptors show that IL-5 signaling, as well as IL-3 and GM-CSF, can occur through receptor heterodimerization, obviating the need for higher-order assemblies of βc dimers. These findings provide insights into IL-5 and βc receptor family signaling mechanisms, aiding in the development of therapies for diseases involving deranged βc signaling.

Keywords: GM-CSF; IL-3; IL-5; common beta; cytokines; signaling.

Figure 1.. Architecture of the IL-5 ternary signaling complex.

(A) Cartoon representation of the components of the hetero-hexameric IL-5 ternary signaling complex. IL-5Rα (teal), IL-5 (grey and blue), βc (red and pale red). (B) Focused reference-free 2D averages from cryoEM of the IL-5 ternary complex. (C) Refined and sharpened cryoEM density maps and ribbon representation of the atomistic modeling of IL-5 ternary complex, colored as in A. (D) The D1, D4 CHR interaction region of βc and IL-5. Hydrogen bonds represented in dashed lines. (E) The D2, D3 CHR interaction region of IL-5Rα and IL-5. (F) The D1 Ig interaction region of IL-5Rα and IL-5. Hydrogen bonds represented in dashed lines.

Figure 2.. Hexameric receptor complexes of the βc family.

(A) Global (top) and focused (bottom) reference-free 2D averages from cryoEM of the IL-5, IL-3, and GM-CSF ternary complexes. (B) Refined and sharpened cryoEM density maps of IL-5, IL-3, and GM-CSF ternary complexes. IL-5Rα (teal), IL-3Rα (yellow), GM-CSFRα (purple), cytokines (grey and blue), βc (red and pale red). (C) Refined and sharpened cryoEM density maps and ribbon representation of the atomistic modeling of βc family ternary complexes, IL-5 (this study), IL-3 (6NMY), GM-CSF (4NKQ/4RS1), colored as in B. Black pointer indicating flexibility in βc.

Figure 3.. Single-molecule imaging of IL-5 receptor complexes.

(A-D) Cartoons illustrating the constructs and labeling conditions used in live cell two color single molecule imaging. IL-5Rα (teal), IL-5 (grey and blue), βc (red and pale red). αALFAtag and αGFP nanobodies used for labeling are depicted in grey. Effective DOL for both nanobodies: ~60%. (E-H) Scatter plots showing relative dimerization levels determined by single-molecule co-tracking for homodimerization and heterodimerization of IL-5Rα and βc in the absence and presence of IL-5, as depicted in A-D. In the absence of IL-5, n=28 for E, n=29 for F, n=25 for G, and n=20 for H. In the presence of IL-5, n=30 for E, n=33 for F, n=34 for G, and n=24 for H. (I,K) Cartoons illustrating the constructs and labeling conditions used in live cell three color single molecule imaging, with the receptor subunits colored as in A-D. (J) Scatter plot showing relative higher order complex formation levels determined by three-color single-molecule co-tracking of IL-5Rα (ctl; n=27, absence of IL-5: n=21, presence of IL-5: n=22, compared to a positive control of a pre-trimerized ALFA-mLepRdel866-tFoldon construct) and (L) βc (ctl; n=26, absence of IL-5: n=23, presence of IL-5: n=36, compared to a positive control of a pre-trimerized XFP-mLepRdel966-tFoldon construct) as depicted in I and K. For E-H, J and L each datapoint represents the result from one evaluated cell. Statistics were performed using two-sample Kolmogorov–Smirnov test (not significant (ns); *P < 0.05 and ****P < 0.0001).

Figure 4.. JAK association and heteromeric receptor signaling in the βc family.

(A) Native heteromeric signaling driven by the IL-5 receptor complex hexamer. IL-5Rα (teal), IL-5 (grey and light grey), βc (red and pale red). (B) Native heteromeric signaling driven by the IL-3 receptor complex hexamer. IL-3Rα (yellow), IL-3 (grey), βc (red and pale red). (C) Native heteromeric signaling driven by the GM-CSF receptor complex hexamer. GM-CSFRα (purple), GM-CSF (grey), βc (red and pale red), hoIL-2Rβ (green), hoIL-2 (orange), γc (blue). (D) Ortho-chimeric heteromeric signaling. IL-5Rα (teal), βc (red and pale red), hoIL-2Rβ (green), hoIL-2 (orange), γc (blue). (E) Ortho-chimeric heteromeric signaling. IL-3Rα (yellow), βc (red and pale red), hoIL-2Rβ (green), hoIL-2 (orange), γc (blue). (F) Ortho-chimeric heteromeric signaling. GM-CSFRα (purple), βc (red and pale red), hoIL-2Rβ (green), hoIL-2 (orange), γc (blue). (G) pSTAT5 signaling of the IL-5 ortho-chimeric heteromeric signaling system depicted in D. (H) pSTAT5 signaling of the IL-3 ortho-chimeric heteromeric signaling system depicted in E. (I) pSTAT5 signaling of the GM-CSF ortho-chimeric heteromeric signaling system depicted in F. Points for G-I were fit to a sigmoidal dose-response model using Prism 9.5.1 (GraphPad). (J) A schematic representation of the live cell JAK association assay. Private receptor (IL-5Rα, IL-3Rα, or GM-CSFRα) in dark grey, common receptor (βc) in red, BFP in blue, HaloTag in grey, HaloTag ligand (HTL) in black, JAK in cyan, and GFP in green. (K) Relative contrast in the GFP channel for experiments depicted in J for the various private (IL-5Rα, IL-3Rα, or GM-CSFRα) and common beta receptor (βc) with different JAK (JAK1, JAK2 and TYK2) FERM-SH2 domains. GFP only was used as a negative control.