Fig 2: E3C4 enables sustained in vivo production of functional bispecific antibodies. (A) Schematic of the single-dose E3C4 administration and blood sampling timeline. (B and C) Serum concentration-time profiles of (B) EpCAM×CD3 and (C) Claudin18.2×41BB bispecific antibodies following a single intraperitoneal dose of E3C4 (3 or 10 μg). (D and E) Functional activity of serum-derived (D) EpCAM×CD3 (T cell-dependent cytotoxicity) and (E) Claudin18.2×41BB (4–1BB reporter activation) compared to recombinant protein standards. (F and G) Pharmacokinetic profiles of both bispecific antibodies over three weeks of repeated E3C4 administration (3 μg, weekly). Data are mean ± SD (N=5). The dotted lines denote the baseline.

Fig 3: E3C4 achieves potent antitumor efficacy by enhancing T-cell immunity with a favorable safety profile in PBMC-humanized mice. (A) Schematic of the treatment schedule. (B) Tumor growth curves and (C) final tumor weights of subcutaneous NCI-N87 xenografts. (D) Quantification of tumor-infiltrating human CD45+ (hCD45+) T cells. Proportions of Granzyme B+ (E) and Ki67+ (F) cells among hCD45+CD8+ T cells. Serum levels of IL-6 (G), IFN-γ (H), and TNF-α (I) measured 24 h after the first dose. Data are presented as mean ± SD. In Figure 4C–I #P < 0.05, ##P < 0.01, ###P < 0.001; *P < 0.05, **P < 0.01 vs EpCAM×CD3 group; ns, not significant (P > 0.05). N = 5 for tumor volume, weight and immune cell and cytokine analysis.

Fig 4: ASP cleaves EpCAM at specific sites within the extracellular domain. (A) His-tagged recombinant EpCAM at a final concentration of 0.15 mg/mL was incubated with increasing concentrations of purified ASP for 3 h. Samples were analyzed by SDS-PAGE and visualized by CBB staining (upper panel), anti-EpCAM immunoblotting (middle panel), and anti-His-tag immunoblotting (lower panel). ASP induced concentration-dependent fragmentation of EpCAM. (B) N-terminal sequencing of ASP-generated fragments a, b, and c yielded the sequences AKPEG, ALQN, and QKEI, respectively. These represent the major resolvable fragments; additional minor cleavage products were observed but could not be isolated for sequencing. (C) Schematic diagram showing the identified ASP cleavage sites (arrows) on the EpCAM extracellular domain. The diagram represents a simplified view of the major cleavage events.

Fig 5: ASP and pro‑aerolysin synergistically enhance bacterial translocation across epithelial monolayers. (A) Time-course analysis of bacterial translocation across epithelial monolayers treated with untreated control (NT), ASP alone, pro‑aerolysin alone, or combined ASP + pro‑aerolysin treatment over 6 h. A. veronii sobria strain 104 (lacking endogenous ASP and aerolysin production) was used to assess translocation. ASP was used at 100 nM; pro-aerolysin was used at 10 nM. The 104 strain was inoculated at MOI 500. Data represent mean ± SEM (n = 3 independent experiments, each performed with an independently prepared T84 monolayer and bacterial inoculum). Statistical analysis was performed using one-way ANOVA, followed by Dunnett’s multiple-comparison test with NT as the control. *P < 0.05, **P < 0.01, and ***P < 0.001. (B) ASP-mediated degradation of junctional and adhesion-related proteins. Representative immunoblots showing the effects of different treatments on tight junction proteins (ZO-1, ZO-2, ZO-3, claudin-7, claudin-3), epithelial adhesion molecule (EpCAM), and GAPDH (loading control) at 3 h post-infection. Upper panels show immunoblot results for each protein under six treatment conditions. Lower panel shows quantitative densitometric analysis of EpCAM degradation fragments expressed as fold-change relative to NT (set as 1.0). Data represent mean ± SEM (n = 3). Statistical analysis was performed using one-way ANOVA, followed by Dunnett’s multiple-comparison test with NT as the control. *P < 0.05, **P < 0.01, and ***P < 0.001.