CTLA-4 antibody-drug conjugate impairs regulatory T-cell and leads to B-cell depletion.

(A) SDS-Reducing gel showing size shift after DM1 conjugation. Lanes (left to right) are 1 hIgGFc, 2 hIgGFc-DM1 antibody-drug conjugate (ADC), 3 Ipilimumab (Ipi), 4 Ipilimumab-DM1 (Ipi-DM1) antibody-drug conjugate (ADC), and 5 Ladder. (B) ELISA binding of hIgGFc, hIgGFc-DM1, Ipi, and Ipi-DM1 to pre-coated 1 μg/mL His-hCTLA-4 and detected with anti-hIgG-HRP (n=2). (C) Flow cytometry binding of Ipi and Ipi-DM1 to hCTLA-4 expressing CHO cells (CHO-hCTLA-4) and detected with anti-hIgG-AF488. (D) Cell viability of CHO-hCTLA-4 cells and wild type CHO cells after 72 hours incubation with Ipi or Ipi-DM1 as measured by MTT assay (n=2). (E) Diagram of experimental design, Ctla4h/h mice were treated intraperitoneally (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 every three days and mice were bled or harvested for bone marrow extraction on day 9 for downstream flow analysis. (F-K) Flow data analysis of day 9 peripheral blood. (F) Gating strategy defining T-cell types, Tregs (CD4+ Foxp3-) and CD4-nonTreg (CD4+ Foxp3-). (G) % Foxp3+ in CD4 and normalized cell number. (H) Relative CTLA-4 Level in Tregs. (I) % Ki67 in Tregs. (J) FACS profile of B cells defined (CD45+ B220+) and data summaries of % B220+ in CD45 and normalized cell number. (K) % Ki67+ in B cells. (L-Q) flow data analysis of day 9 bone marrow after gating on CD45+ B220+, mature (M) (IgDhi IgMlow), immature transitional type 1 (T1) (IgDlow IgM+), and Pre-pro/Pro/Pre B cell subtypes (B220+ IgM-). (L) FACS profile gating, (M) % of mature B cells in B220 and absolute cell number summaries, (N) % of T1 immature B cells in B220 and absolute cell number summaries. (O) FACS profile gating of Pre-pro/Pro/Pre B cells, (P) % of Pre-pro/Pro/Pre B cells in B220 and absolute cell number summaries. (Q) CD21/35 expression in mature B cells in bone marrow. (B-D) representative data two or more repeats. (F-K) Data combined from two independent experiments (n=13-14). (L-Q) Data representative of two independent experiments (n=5). Data analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

B-cell depletion is transient and correlated with a decrease in regulatory T-cell.

(A) Diagram of experimental design, Ctla4h/h mice were treated intraperitoneal (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 every three days and mice were bled on day 3,9,13, 25 for downstream flow analysis. (B-F) Flow data of time course of CD45, B, and Treg cells in peripheral blood, (B) CD45 normalized cell number, (C) % of B-cells in CD45, (D) B-cell normalized cell number, (E) % Foxp3+ Tregs in CD4, (F) Treg normalized cell number. (G-J) Flow analysis of B-cell subtypes from day 13 peripheral blood. (G) FACS profile of Mature (M) (IgDhi IgMlow), and immature Transitional type 1 (T1) (IgDlow IgM+) B cell subtypes after gating on CD45+ B220+ B cells. (H) % of mature B cells in B220. (I) % of T1 B cells in B220. (B-F) Data is combined from two independent experiments (n=10). (G-I) Data is representative of two independent experiments (n=6). Data were analyzed by an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001

CTLA-4 antibody-drug conjugate leads to T-cell activation.

Flow data analysis of peripheral blood from Ctla4h/h mice on day 9 after treatment with hIgGFc or ADC (hIgGFc-DM1 or Ipi-DM1). (A) % CD4+ in CD45 and normalized cell number. (B) % CD8+ in CD45 and normalized cell number. (C) % Ki67+ in CD4 and CD8 T cells. (D, E) FACS profiles and summaries depicting the increase in effector T-cells after Ipi-DM1 treatment, naïve (Q1: CD44 CD62Llow), central memory (Q2: CD44low CD62Lhi) and effector (Q3: CD44hi CD62Lhi), (D) phenotype of CD4+ T cells (E), and CD8+ T cells (H). (B-C) Mice treated with hIgGFc or Ipi-DM1, data combined from two independent experiments (n=13-14). (D-E) Mice treated with hIgGFc or Ipi-DM1, data is representative of two independent experiments (n=5). Data analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. Non-significant [ns], *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

B-cell depletion depends on T-cells but not macrophage.

(A) Diagram of experimental design, male or female Ctla4h/h mice were treated intraperitoneal (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 with/out (T-cell depleting antibody 100 μg/mouse of anti-Thy1.2 or 150μL/mouse of clondrosome or depleting antibody CD4 or CD8 100 μg/mouse) every three days and mice were bled on day 9 for downstream flow analysis. (B) FACS profiles depicting gating strategy after gating on CD45 for CD8 & CD4 T (top panel) and B cells (bottom panel). (C-E) Data summaries from FACS profile(panel B), (C) % CD4+ in CD45 and normalized cell number, (D) % CD8+ in CD45 and normalized cell number, and (E) % B220+ in CD45 and normalized cell number. (F) FACS profiles depicting gating strategy after gating on CD45 for CD8 & CD4 T (top panel) and B cells (bottom panel). (G-I) Data summaries from FACS profile(panel F), (G) % CD4+ in CD45 and normalized cell number, (H) % CD8+ in CD45 and normalized cell number, and (I) % B220+ in CD45 and normalized cell number. (B-E) Data combined from two independent experiments (n=10) and analyzed by ordinary one-way ANOVA with Tukey’s multiple comparisons test and represented as mean ± SEM. Non-significant [ns], *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. (F-I) Data combined from two independent experiments (n=11) and analyzed by unpaired two-tailed Student’s t test and represented as mean ± SEM. Non-significant [ns], *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Mutant soluble CTLA-4-Ig rescues B-cell.

Ctla4h/h mice were treated intraperitoneal (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 with/out (100 μg/mouse of Belatacept) every three days for total of three doses and mice were bled on day 9 for downstream flow analysis.(A, D) FACS profiles depicting gating strategy after gating on CD45 for (A) Tregs (% Foxp3 in CD4+ T cells) and (D) B cells (% B220+ in CD45). (B) % Foxp3+ (Tregs) in CD4 and normalized cell number. (C) Representative FACS profile of CTLA-4 expression in Tregs (CD4 Foxp3) and Relative CTLA-4 summary. (E) % B220+ in CD45 and normalized cell number. (F-G) FACS gate and summaries showing mutant CTLA-4-Ig can decrease effector memory T-cells associated with Ipi-DM1 treatment, naïve (Q1: CD44low CD62L), central memory (Q2: CD44 CD62L) and effector (Q3: CD44 CD62L), phenotype of CD4 T cells (F), and CD8 T cells (G). (H) FACS of GranzymeB gating in CD45 (Top panel), CD4 (middle panel), and CD8 (bottom panel). (I-K) GranzymeB expression summaries in (I) CD45, (J) CD4, and (K) CD8 cells. Data (A-E) combined from three independent experiments (n=15). Data (F-K) are representative of two independent experiments (n=5). Data (A-K) were analyzed by ordinary one-way ANOVA with Tukey’s multiple comparisons test and represented as mean ± SEM. Non-significant [ns], *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

B-cell depletion is partially rescued by anti-TNF-alpha.

(A) ELISA binding of clinical grade drug Adalimumab (Humira) to pre-coated 1μg/mL mouse-TNF-alpha and detected with anti-hIgG-HRP, Ipilimumab negative control. (B) Diagram of experimental design, male Ctla4h/h mice were treated intraperitoneal (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 with/out (100 μg/mouse of Adalimumab) every three days for total of three doses and mice were bled on day 9 and 13. (C) FACS profiles depicting gating strategy after gating on CD45 for Tregs (% Foxp3 in CD4) (top panel) and B cells (% B220+ in CD45) (bottom panel) from day 9 peripheral blood. (D) % Foxp3+ (Tregs) in CD4 and normalized cell number. (E) % B220+ in CD45 and normalized cell number. (F, G) Day 13 peripheral blood post red blood lysis by ACK buffer were cultured and stimulated in the presence of Iononmycin/PMA, and GolgiPlug for 4 hrs followed by intracellular cytokine detection (TNF-alpha, IFN-gamma) in CD4 and CD8 T cells, (F) TNF-alpha, and (G) IFN-gamma. Data (C-E) are combined from two independent experiments (n=11-12) and analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001. Data (F, G) is representative of two independent experiments (n=5-6) and analyzed by ordinary one-way ANOVA with Tukey’s multiple comparisons test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

CD4-nonTregs and CTLA-4 expression.

(A) % CD4 Foxp3 in CD45 and normalized cell number. (B) Relative CTLA-4 level in CD4-nonTregs. Data combined from two independent experiments (n=13-14) and analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

B-cell depletion is not mediated by DM1 payload off-target release.

(A) % Foxp3 in CD4 and normalized cell number. (B) Relative CTLA-4 level in Tregs. (C) % Ki67 in Tregs. (D) % CD4 Foxp3 in CD45 and normalized cell number. (E) Relative CTLA-4 level in CD4-nonTregs. (F) FACS profiles depicting gating strategy after gating on CD45 and data summaries of % B220 in CD45 and normalized cell number. (G) % Ki67 in B cells. Data combined from two independent (n=11) and analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

B-cells do not express CTLA-4.

FACS profile of human CTLA-4 in wild-type and human CTLA-4 knock-in mice peripheral blood, (A) % CTLA-4+ in Tregs (CD45+ CD4+ Foxp3+), and (B) % CTLA-4+ in B cells (CD45+ B220+). Data representative of 2 mice.

IgM negative B-cell subtypes in bone marrow.

(A) FACS profile gating of Pre B (IgM-CD43-) and Pre-pro/Pro (IgM-CD43+) cells in B220. (B-C) % of B cell subtype in B220 and absolute cell numbers, (B) Pre B cells, (C) Pre-pro/Pro B cells. Data is representative of one experiment (n=5) and analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

T-cell quantity and proliferation are not impacted by IgG-DM1 treatment.

(A) % CD4 in CD45 and normalized cell number. (B) % CD8 in CD45 and normalized cell number. (C) % Ki67 in CD4 and CD8 T cells. Mice treated with hIgGFc or hIgGFc-DM1, data combined from two independent experiments (n=11) and analyzed using an unpaired two-tailed Student’s t test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Mutant soluble CTLA-4-Ig does not neutralize Ipilimumab or its drug-conjugate.

(A) Detached CHO cells expressing hCTLA-4 were incubated with serial dilution of Ipi or Ipi-DM1 in the presence of given dose of Abatacept or Belatacept on ice for 30 minutes followed by FACS detection using mouse anti-human IgG AF488, mean fluorescence intensity (MFI). (B) MTT cell viability of CHO-hCTLA-4 cells after 72 hours incubation with Ipi or Ipi-DM1 in the presence of given dose of Abatacept or Belatacept. Data is representative of two independent experiments (n=3).

T-cell cytokine production.

Day 13 peripheral blood samples from treated (hIgGFc or Ipi-DM1 with/out Belatacept) mice post ACK buffer red blood lysis were cultured and stimulated in the presence of Iononmycin/PMA, and GolgiPlug for 4 hrs followed by intracellular cytokine detection (TNF-alpha, IFN-gamma) in CD4 and CD8 T cells. Data is representative of two independent experiments (n=5) analyzed by ordinary one-way ANOVA with Tukey’s multiple comparisons test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Anti-FASL fails to rescue B cell.

Ctla4 mice were treated intraperitoneally (i.p.) (100 μg/mouse) with hIgGFc or Ipi-DM1 with/out (100 μg/mouse of anti-FASL) every three days for a total of three doses and mice were bled on day 9. (A) FACS profile after gating on CD45 for B cells. (B) % B220 in CD45 and absolute B cell number summaries. Data is representative of one experiment (n=5). Data analyzed by ordinary one-way ANOVA with Tukey’s multiple comparisons test and represented as mean ± SEM. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

Drug to antibody ratio (DAR)