Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: a, Structure of WT IL-2 model, based on Protein Data Bank (PDB) ID 1M47 (cartoon representation), bound to IL-2Rαβγ from PDB ID 2B5I; IL-2Rα (CD25) shown in gray (cartoon representation), IL-2Rβ (CD122) and IL-2Rγ (CD132) shown in light gray (surface representation). Rosetta Design was used to replace the regions in red with short, structured loops (light blue) that reconnect α-helices H1-H4, generating relooped designs R1.1-38. b, IL-2Rβγ signaling activity (HEK-Blue reporter assay) and expression level from Expi293 supernatants diluted 1:400; designs R1.1-38 normalized to WT IL-2 levels. c, Optimization of design R1.1 by extending IL-2Rγ-interacting helices and redesigning the H2-H3 loop (“Round 2 optimization”) to produce design R2.1, which was further optimized by mutating two IL-2Rβ-interacting residues, R88N and I99L, to produce design R2.2. d, Binding affinity to IL-2Rβγ-Fc and IL-2Rα as measured by biolayer interferometry (BLI): design R2.2 binds IL-2Rβγ with a K D close to that of WT IL-2 (0.80 nM and 0.44 nM, respectively), consistent with previously reported measurements for WT IL-2 , . The purified proteins measured here contain STII tags; NB = no binding. e, Circular dichroism (CD) wavelength scan at 25℃ ( left ) and thermal melt monitoring helical signal at 208 nm ( right ). f, STAT5 phosphorylation (pSTAT5) in resting T cell subtypes after treatment with the indicated designed cytokines or WT IL-2. NC = not calculated.

Article Snippet: To measure interactions between designed cytokines and IL-2Rβγ, Fc-tagged IL-2Rβγ (Acro Biosystems) or biotinylated Fc-tagged IL-2Rβγ (Acro Biosystems) were captured on ProA or SA biosensors to ∼3 nm (SA biosensors were used for all VHH fusion proteins due to VHH binding to ProA); binding was measured with analyte serially diluted from 100 to 0.4 nM, using 120 s association and 480 s dissociation.

Techniques: Activity Assay, Reporter Assay, Expressing, Binding Assay, Purification, Circular Dichroism, Phospho-proteomics

Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: Designs were expressed in Expi293 cell and supernatants collected and evaluated for binding to IL-2Rα by ELISA ( top , normalized to WT IL-2) and IL-2Rβγ signaling activity by HEK-Blue assay ( bottom , levels normalized to that of WT IL-2). NTC is a non-transfected control.

Article Snippet: To measure interactions between designed cytokines and IL-2Rβγ, Fc-tagged IL-2Rβγ (Acro Biosystems) or biotinylated Fc-tagged IL-2Rβγ (Acro Biosystems) were captured on ProA or SA biosensors to ∼3 nm (SA biosensors were used for all VHH fusion proteins due to VHH binding to ProA); binding was measured with analyte serially diluted from 100 to 0.4 nM, using 120 s association and 480 s dissociation.

Techniques: Binding Assay, Enzyme-linked Immunosorbent Assay, Activity Assay, Transfection, Control

Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: a, Schematic of OUTSMART dIL-2 system. In response to antigen-stimulated T cell activation, engineered T cells secrete the dIL-2 designed cytokine, which is targeted to CD8α to favor the engagement of IL-2Rβγ on endogenous NK and T cells in addition to the engineered T cells. b, dIL-2 is encoded in a single lentiviral vector that is broadly compatible with engineered CAR-T and TCR-T therapies. The dIL-2 protein is expressed under the control of a regulated promoter, and consists of c, Designed cytokine R2.2 (teal) fused to an anti-CD8α VHH (blue) via a 3xG4S linker. The VHH shown is a model for illustrative purposes (there is no crystal structure of the VHH used in this work). CD8α (PDB ID 8EW6) and IL-2Rβγ receptors (PDB ID 2B5I) are shown in surface representation. d, pSTAT5 activity in resting and activated T cell subtypes after treatment with recombinant protein of the indicated cytokines. e, Schematic of regulated promoter screen. Concatenated enhancer motifs were engineered into CAR lentiviral backbones utilizing either a dual-promoter (constitutive MND CAR with inducible cytokine) or single-promoter (CAR and cytokine expressed from the same inducible promoter) architecture. CAR-T cells were evaluated for inducible cytokine production in the presence or absence of ROR1 antigen. f, Regulated promoter-driven cytokine production at rest (unstimulated) versus after antigen stimulation. Illustrations in a were created using BioRender ( https://biorender.com , Fontana, T. 2024).

Article Snippet: To measure interactions between designed cytokines and IL-2Rβγ, Fc-tagged IL-2Rβγ (Acro Biosystems) or biotinylated Fc-tagged IL-2Rβγ (Acro Biosystems) were captured on ProA or SA biosensors to ∼3 nm (SA biosensors were used for all VHH fusion proteins due to VHH binding to ProA); binding was measured with analyte serially diluted from 100 to 0.4 nM, using 120 s association and 480 s dissociation.

Techniques: Activation Assay, Plasmid Preparation, Control, Activity Assay, Recombinant

Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: a, Binding affinity to IL-2Rβγ-Fc and CD8α measured by biolayer interferometry (BLI) for design R2.2 (untargeted), dIL-2 protein (R2.2 fused to anti-CD8α VHH), and anti-CD8α VHH alone. b, Representative flow cytometry plots from resting ( left ) and activated ( right ) CD8 T cells (corresponds to ) measuring pSTAT5 (Alexa Fluor 647) signal after treatment with the indicated cytokine at a final concentration of 0.065 nM. dIL-2 exhibits superior pSTAT5 of resting T cells and similar pSTAT5 of activated T cells compared to WT IL-2. CD8 T cells are defined by gating on CD3+CD4-signal. c, Compared to WT IL-2, dIL-2 exhibits less pSTAT5 in resting CD4+Foxp3-cells after treatment with the indicated cytokine. d , Fold proliferation of T cells that were rested or stimulated for 24h with anti-CD3/anti-CD28 (TransAct) before incubation with indicated recombinant cytokines at 0.065 nM; cells were counted and split into fresh cytokine media every 3-4 days. e , Frequency of CD8+ cells (defined as CD3+CD4-) in cells stimulated for 24h and grown with the indicated recombinant cytokines at 0.065 nM. By day 14, 81.3% of cells were CD8+ after addition of dIL-2 protein. f , GS linker does not impact dIL-2 protein function: pSTAT5 elicited by dIL-2 protein with or without 3XG4S linker between the CD8ɑ binder and cytokine elements; representative pSTAT5 staining in CD8+ T cells treated with 0.065 nM cytokine ( left ), and pSTAT5 in resting T cell subtypes after treatment with the indicated designed cytokines ( right ).

Article Snippet: To measure interactions between designed cytokines and IL-2Rβγ, Fc-tagged IL-2Rβγ (Acro Biosystems) or biotinylated Fc-tagged IL-2Rβγ (Acro Biosystems) were captured on ProA or SA biosensors to ∼3 nm (SA biosensors were used for all VHH fusion proteins due to VHH binding to ProA); binding was measured with analyte serially diluted from 100 to 0.4 nM, using 120 s association and 480 s dissociation.

Techniques: Binding Assay, Flow Cytometry, Concentration Assay, Incubation, Recombinant, Staining

Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: a, Structure of WT IL-2 model, based on Protein Data Bank (PDB) ID 1M47 (cartoon representation), bound to IL-2Rαβγ from PDB ID 2B5I; IL-2Rα (CD25) shown in gray (cartoon representation), IL-2Rβ (CD122) and IL-2Rγ (CD132) shown in light gray (surface representation). Rosetta Design was used to replace the regions in red with short, structured loops (light blue) that reconnect α-helices H1-H4, generating relooped designs R1.1-38. b, IL-2Rβγ signaling activity (HEK-Blue reporter assay) and expression level from Expi293 supernatants diluted 1:400; designs R1.1-38 normalized to WT IL-2 levels. c, Optimization of design R1.1 by extending IL-2Rγ-interacting helices and redesigning the H2-H3 loop (“Round 2 optimization”) to produce design R2.1, which was further optimized by mutating two IL-2Rβ-interacting residues, R88N and I99L, to produce design R2.2. d, Binding affinity to IL-2Rβγ-Fc and IL-2Rα as measured by biolayer interferometry (BLI): design R2.2 binds IL-2Rβγ with a K D close to that of WT IL-2 (0.80 nM and 0.44 nM, respectively), consistent with previously reported measurements for WT IL-2 , . The purified proteins measured here contain STII tags; NB = no binding. e, Circular dichroism (CD) wavelength scan at 25℃ ( left ) and thermal melt monitoring helical signal at 208 nm ( right ). f, STAT5 phosphorylation (pSTAT5) in resting T cell subtypes after treatment with the indicated designed cytokines or WT IL-2. NC = not calculated.

Article Snippet: The activity of relooped IL-2 molecules in Expi293 supernatants (undiluted, and diluted 1:10, 1:50, 1:100, 1:200, and 1:400) was assayed on HEK-Blue IL2R cells (Invivogen) according to the manufacturer’s instructions.

Techniques: Activity Assay, Reporter Assay, Expressing, Binding Assay, Purification, Circular Dichroism, Phospho-proteomics

Journal: bioRxiv

Article Title: A Dynamically Regulated Designed IL-2 for Tumor-localized Signaling

doi: 10.64898/2026.09.18.752703

Figure Lengend Snippet: Designs were expressed in Expi293 cell and supernatants collected and evaluated for binding to IL-2Rα by ELISA ( top , normalized to WT IL-2) and IL-2Rβγ signaling activity by HEK-Blue assay ( bottom , levels normalized to that of WT IL-2). NTC is a non-transfected control.

Article Snippet: The activity of relooped IL-2 molecules in Expi293 supernatants (undiluted, and diluted 1:10, 1:50, 1:100, 1:200, and 1:400) was assayed on HEK-Blue IL2R cells (Invivogen) according to the manufacturer’s instructions.

Techniques: Binding Assay, Enzyme-linked Immunosorbent Assay, Activity Assay, Transfection, Control