Fig 1: In vitro cytokine production scales with programmed chimeric antigen receptor (proCAR) oligomeric state including tetramers.(a) SDS-PAGE migration of programmed membrane protein (proMP) C4.1 is consistent with a tetramer. Design model and peptide sequence shown for reference. (b) Rosetta ab initio structure prediction calculations predict that proMP C4.1 preferentially forms a tetramer. (c) CAR surface expression on primary mouse CD8+ T cells stably expressing CD28TM and proCAR-4 analyzed by c-Myc staining on flow cytometry. HER2 proCAR-4 was designed using the proMP C4.1 sequence without the final C-terminal leucine as a transmembrane domain (TMD), inserted as shown in Figure 3a. (d) IncuCyte killing assay over 24 hr of no CAR, CD28TM, and proCAR1-4 T cells on MC57-HER2 target cells at 1:1 effector to target ratio. Comparison of maximum killing for n = 6 independent experiments shown between CD28TM vs. ProCAR-4. Data points represent individual experiments, with mean ± SEM error bars plotted. (e) Tumor growth over time using the same experimental design in Figure 5a for No CAR (empty vector), CD28TM WT, and proCAR-4 T cell groups (n = 5–6 mice/group). Data points represent mean ± error bars showing SEM. Statistical analysis performed using a two-way ANOVA at day 10. (f, g) Linear correlation of proCAR oligomeric state vs. IFN?, IL-2, TNFa, and GM-CSF cytokine production (normalized to CD28TM reference) from 24 hr co-culture with (e) MC57-HER2 and (f) antigen-negative MC57 tumor cells. Individual data points are colored, mean values in white box and error bars indicate SEM.