Fig 2: F5111 IC blocks IL-2 binding to IL-2Rß and biases toward Treg activation(A) Equilibrium biolayer interferometry-based titrations of F5111 antibody (Ab), hIL-2, control IC, hIL-2/F5111 complex (Cx, 1:1 molar ratio), and F5111 IC LN35 binding to hIL-2 (left), hIL-2Ra (middle), or hIL-2Rß (right).(B) STAT5 phosphorylation response of IL-2Ra+ (left) and IL-2Ra- (right) YT-1 cells stimulated with hIL-2, control IC, hIL-2/F5111 Cx (1:1 molar ratio), or F5111 IC LN35.(C) STAT5 phosphorylation responses of stimulated human Treg (left), CD8+ T (middle), and TConv (right) cells. Data represent mean ± SD (n = 3). See also Figures S1 and S2; Tables S2, S3, and S4.

Fig 3: F5111 IC expands functional Tregs without compromising infection immunity(A–C) C57BL/6 mice (n = 3 per group) were treated daily for 4 days (intraperitoneally) with PBS or 8.2 µg F5111 IC (1.5 µg IL-2 equivalence) and spleens were harvested at the indicated times.(A) Representative flow plots showing Treg percentage and IL-2Ra versus FOXP3 levels in CD4+ T cells at the indicated times. (B) Percent Tregs within CD4+ T cells at the indicated times. (C) Treg:CD8+ T (left), Treg:TConv (middle), and Treg:NK (right) at the indicated times. Dashed lines show baseline values for the PBS-treated cohort harvested 1 day after the last dose.(D) Serum half-life of F5111 IC in C57BL/6 mice (n = 5) treated retro-orbitally with 2 mg/kg F5111 IC (~0.4 mg/kg IL-2 equivalence).(E) C57BL/6 CD45.1 RFP-FOXP3 mice were treated daily for 4 days with PBS (n = 3) or 6.2 µg F5111 IC (1.125 µg IL-2 equivalence, n = 2). Percent proliferation suppression of TConv cells from untreated C57BL/6 CD45.2 mice (n = 5) by titrating ratios of Tregs from spleens of treated mice is shown.(F) C57BL/6 mice (n = 5 per group) were dosed on days 0, 2, and 4 with PBS or 8.2 µg (1.5 µg IL-2 equivalence) control IC or F5111 IC. Spleens were harvested on day 6. Number of cTregs (CD4+FOXP3+IL-2RaHighBCL-2High) and effector (eTregs) (CD4+FOXP3+IL-2RaLowBCL-2Low) are shown.(G–I) C57BL/6 mice were administered 25 cysts of T. gondii on day 0. The control group was not given cysts. Starting on day 1, mice were treated daily for 5 days with PBS (control, n = 5; PBS, n = 4) or 8.2 µg (1.5 µg IL-2 equivalence) control IC (n = 5) or F5111 IC (n = 5). Mice were sacrificed on day 10. (G) Mouse weight. (H) Parasite burden. (I) Treg:T-bet+Tetramer (Tet)+ CD8+ T cell ratio in the spleen. (G–I) Mean ± SEM. All other data are mean ± SD. Statistical significance in percent initial weight on day 10 compared with F5111 IC is shown. *p = 0.05, **p = 0.01, ***p = 0.001, ****p = 0.0001. See also Figure S5.

Fig 4: F5111 IC shows greater Treg bias than hIL-2/F5111.2 complex(A) pSTAT5 response of stimulated human Treg (left), CD8+ T (middle), and TConv (right) cells. hIL-2/F5111.2 complex (Cx) was at a 1:1 molar ratio.(B–E) NOD mice (n = 4 per group) were treated daily for 4 days with PBS, 1.5 µg hIL-2 with 6.6 µg F5111.2 antibody (1:2 molar ratio), or 8.2 µg (1.5 µg IL-2 equivalence) control IC or F5111 IC.(B) Number of Treg, CD8+ T, TConv, and NK cells.(C) MFI of FOXP3 within Tregs (left) and IL-2Ra MFI within Treg, CD8+ T, and TConv cells.(D) Treg:CD8+ T (left), Treg:TConv (middle), and Treg:NK cell (right) ratios.(E) Percent Tregs within the CD4+ T cell population.(F) Percent Ki-67+ cells within Treg, CD8+ T, TConv, and NK cells from NOD mice (n = 5 per group) treated as in (B–E). Data represent mean ± SD. Statistical significance between control IC, hIL-2/F5111.2 Cx, and F5111 IC shown. *p = 0.05, **p = 0.01, ***p = 0.001, ****p = 0.0001. See also Figure S5; Table S4.

Fig 5: Tuning IC intramolecular affinity modulates Treg bias(A) hIL-2/F5111 crystal structure (PDB: 5UTZ) with alanine-mutated residues shown in yellow (LC) or green (HC). Human IL-2Ra is overlaid from the IL-2 quaternary complex structure (PDB: 2B5I).(B) Equilibrium biolayer interferometry-based titrations against hIL-2Rß.(C) STAT5 phosphorylation responses of stimulated human Treg (left), CD8+ T (middle), and TConv (right) cells.(D) Schematic of multivalent binding model (left). Predicted Treg:CD8+ T (middle) and Treg:TConv (right) pSTAT5 ratios for ICs at 10 pM concentration are plotted against predicted IL2Rß/?c KD (nM).(E) Predicted (lines) and experimental (points) percent control IC pSTAT5 EMax on CD8+ T (left) or TConv (right) cells plotted against the predicted (lines) or experimental (points) pSTAT5 EC50 on Tregs for each IC. See also Figure S3; Tables S2 and S4.