CAR signaling domain cut to 45 amino acids: Canelé's agent-designed receptor beats BBζ on surface expression
Synopsis
Canelé used an AI-agent-orchestrated rational design loop to refactor the CAR intracellular signaling domain, compressing the second-generation BBζ domain into a 54-amino-acid design called Stoat and then a 45-amino-acid variant, Happy Stoat. At the same mRNA dose, Happy Stoat puts more CAR on the T-cell surface than both BBζ and Stoat and releases more cytokine. Across five donors, Happy Stoat matched BBζ's B-cell depletion at lower mRNA doses.
Interpretation
A CAR's intracellular signaling domain can be far shorter than current second-generation designs without losing immune activation. CAR engineering has concentrated on the extracellular binder, while signaling domains largely follow a few established designs, such as BBζ, which combines 4-1BB costimulation with CD3ζ activation. A design executed by Opus 4.6 is just 54 amino acids, one-third the length of BBζ, and released more IFN-γ than BBζ while matching its killing.
Compressing further to the 45-amino-acid Happy Stoat moved the advantage onto surface expression. Surface expression was identified as Happy Stoat's key advantage rather than sequence length alone. At the same mRNA dose, Happy Stoat expresses more CARs on the T-cell surface than both BBζ and Stoat, with stronger cytokine release across mRNA doses and E:T ratios.
The expression advantage carries into killing and persistence. BBζ reached on the first day an expression level Happy Stoat approached after one week. In a time-course experiment Happy Stoat maintained high expression far longer than BBζ, and it kills at lower E:T ratios than BBζ, with the gap accentuated a few days after transfection.
The compressed domain shows a dose advantage at the donor level. The comparison remains against the second-generation BBζ construct. Across five donors, Happy Stoat achieved comparable B-cell depletion at lower mRNA doses than BBζ.
Perspective
The demonstrated scope is in vitro and ex vivo: surface expression on T cells, cytokine release, killing and E:T-ratio comparisons for the synthetic constructs, plus B-cell depletion across five donors; in vivo delivery is still being pursued.
The authors describe Happy Stoat as a glimpse of a vast, under-explored sequence space and say they continue to validate and optimize these refactored signaling domains; in vivo results are not yet reported. For the antigen-combination work with Anthropic, the authors expect experimental results within six months.