Journal: bioRxiv

Article Title: AI-Guided CAR Designs and AKT3 Degradation Synergize to Enhance Bispecific and Trispecific CAR-T Cell Persistence and Overcome Antigen Escape

doi: 10.1101/2025.06.12.658477

Figure Lengend Snippet: (A) In-silico analysis of CAR-T cell-treated patients (n=4,219) revealed a high relapse rate, with 42.11% (n=216 of n=513 overall relapse patients) experiencing CD19-negative recurrence after monospecific CAR-Therapy (n=2,916). (B) Schematic overview of the CAR design strategy showing mono, bi, and trispecific constructs targeting CD19, CD20, and CD22. (C) Experimental workflow illustrating CAR screening: 1,452 CARs were transduced into primary T cells and analyzed for signal-1 (activation), signal-2 (exhaustion), and signal-3 (cell death). (D) Categorization of CARs into low (L), medium (M3), and high (H) levels based on fluorescence intensity cutoffs determined by monospecific CD19 CARs. (E) Distribution of 1,452 screened CARs across L-, M-, and H-CARMSeD categories using the CAR-Mediated Self-Destruction (CARMSeD) scoring system. (F) AI model development pipeline for CAR dysfunction risk prediction, based on 1,452 CAR constructs with an 80:20 split for training and testing. (G–J) Performance metrics of AI model predicting CAR-Mediated Self-Destruction (CAR-MSED) scores using 1,452 CAR constructs (G) Model accuracy over 50 epochs, achieving a training accuracy of 0.98 and validation accuracy of 0.95. (H) Scatter plot comparing measured versus predicted CAR-MSED scores for training (R 2 = 0.87) and validation (R 2 = 0.83) sets. (I) Predicted versus measured CAR-MSED scores on the validation set, categorized into low (L-CARMSED, blue), medium (M-CARMSED, orange), and high (H-CARMSED, green) groups. (J) Box plot of predicted signal scores for 9,372 unknown sequences, classified as L-CARMSED (2,749 sequences), M-CARMSED (1,468 sequences), and H-CARMSED (5,155 sequences). (K) Molecular dynamics simulation of CAR constructs with varying linker lengths, assessing CAR-CAR interaction. Structural conformations at 0 ns, 50 ns and 200 ns for different CAR scFv arrangements highlighting CDR regions (surface transparency 30%), Root Mean Square Deviation (RMSD) plots over 200 ns for the both constructs, respectively, indicating structural stability and conformational changes. (L) In vitro receptor binding affinity validation for top humanized scFvs of CD19, CD20, and CD22 CARs (n=6).

Article Snippet: Briefly, CD19, CD22 CAR expression was evaluated using CD19 and CD20 CAR detection antibodies and CD22 CAR expression (Miltenyi Biotec) was evaluted using Protein L-APC (Cell signaling) followed by PE-conjugated anti-biotin secondary antibodies (Miltenyi Biotec).

Techniques: In Silico, Construct, Activation Assay, Fluorescence, Biomarker Discovery, In Vitro, Binding Assay

Journal: bioRxiv

Article Title: AI-Guided CAR Designs and AKT3 Degradation Synergize to Enhance Bispecific and Trispecific CAR-T Cell Persistence and Overcome Antigen Escape

doi: 10.1101/2025.06.12.658477

Figure Lengend Snippet: (A) Schematic illustration of the K562 cell line model expressing individual or triple combinations of CD19 (purple), CD20 (yellow), and CD22 (red) antigens. (B) Bar chart depicting the percentage expression of each antigen in K562 cell lines, both individually and in combination. (B) Cytotoxicity assays showing potent and antigen-specific killing of K562 target cells. All tested constructs surpassed the performance of second-generation monospecific CD19 (m19) CAR-T cells (n=5). (C) Comparison of proliferation rates for bispecific; b20/19 or b22/19, and trispecific; t20/19/22 CAR-T cells. Trispecific constructs showed reduced proliferation, consistent with increased structural rigidity predicted by CARMSeD scoring. (D) Schematic of the Raji WT cell line platform expressing CD19 (purple), CD20 (yellow), and CD22 (red) antigens, edited using CRISPR-Cas9 to generate knockout variants. (E) Cytotoxicity assays demonstrating the superior efficacy of b20/19 CAR-T cells in eliminating antigen-negative Raji variants, compared to ineffective m19 CARs (n=5). (F) Schematic representation of the tumor rechallenge (TR) model using the Raji WT cell line (Raji WT ). Gray circles represent initial engraftment and monitoring phases, while purple circles indicate the timing of the Raji CD19-/- rechallenge. (G) Heatmap representation of TR model showing IFN-γ secretion (pg/mL), percentage of tumor lysis, and the number of CAR-T cells detected on days 7, 9, 11, 15, and 17 post-rechallenge (n=5). (H) Schematic timeline of in vivo lymphoma model for evaluation of monospecific and bispecific CAR-T cells. Mice were xenografted with Raji WT cells (expressing CD19, CD20, and CD22) (day 0), followed by administration of m19 or b20/19 CAR-T cells on day 5 and subsequent Raji CD19-/- TR on day 12, 19 and 26. (I–J) Bioluminescent imaging and tumor burden quantification show effective tumor control by b20/19 CAR-T cells versus m19 CARs. (K) CAR-T cell persistence over time. (L) Kaplan-Meier survival curves showing survival outcomes over 70 days (n=5). (M) Analysis of residual tumor CD19 or CD20 tumor cells over time. (N, O) Bar plot showing Granzyme B and IFN-γ secretion from CD8 + CAR-T cells isolated post-treatment with m19 and b20/19 confirm functional cytotoxicity of b20/19 against CD19⁻ targets (n=5). (P–Q) Repeated TR induced upregulation of exhaustion markers PD-1 and LAG-3 (n=5). (R) Immunophenotyping of CAR-T cells post-TR shows loss of central memory (T cm ) populations and increased PD-1 expression, consistent with functional exhaustion and limited persistence (n=5). Data represents mean ± SEM. *p < 0.05; **p < 0.01; ***p < 0.005; ****p < 0.001. A non-parametric t-test was used for statistical analysis between groups

Article Snippet: Briefly, CD19, CD22 CAR expression was evaluated using CD19 and CD20 CAR detection antibodies and CD22 CAR expression (Miltenyi Biotec) was evaluted using Protein L-APC (Cell signaling) followed by PE-conjugated anti-biotin secondary antibodies (Miltenyi Biotec).

Techniques: Expressing, Construct, Comparison, CRISPR, Knock-Out, Lysis, In Vivo, Imaging, Control, Isolation, Functional Assay

Journal: bioRxiv

Article Title: AI-Guided CAR Designs and AKT3 Degradation Synergize to Enhance Bispecific and Trispecific CAR-T Cell Persistence and Overcome Antigen Escape

doi: 10.1101/2025.06.12.658477

Figure Lengend Snippet: (A) Pathway analysis of proteins involved in AKT3 interaction, modifications or regulation of its expression with emphasis on FOXO4. (B) Relative mRNA expression levels (normalized to beta actin) of key genes show upregulation of FOXO4 mRNA in b20/19-AKT3 PROTAC CAR-T. (C1) Flow cytometry histograms of total FOXO4 and phosphorylated FOXO4 (p-FOXO4) in CAR-T cells after TR with Raji CD19-/- cells (C2) Histogram analysis of the flow cytometry plots (n=10). (D) Bar graph shows the percentage of CD8 + CAR-T cells expressing different phenotypes. Pie charts illustrate the proportional distribution of these subsets across conditions. (E) Persistence of CAR-T cells over 15 days under various conditions (n=4). (F) Violin plots show the percentage of mTOR activity (% mTOR activity) in various conditions, with shRNA based FOXO4 knockdown significantly elevated mTOR activity (n=6). (G) Bar plots show the percentage of MFI of autophagy from autophagic flux assay (n=8). (H) ECAR in NTP PROTAC+Scram , NTP PROTAC+shFOXO4 , AKT3 PROTAC+Scram , and AKT3 PROTAC+shFOXO4 conditions, with FOXO4 knockdown increasing shift from oxidative phosphorylation (OXPHOS) to glycolysis (n=12 data points). (I) Similarly, OCR with FOXO4 knockdown decreasing mitochondrial respiration. Individual data points are shown for each condition (n=12 data points). (J1) Percentage of expression (% expression) of CD19 (yellow), CD20 (blue), and CD22 (purple) across 129 ALL patient samples, with varying expression levels for each marker. (J2) Bar graph displays the number of patient samples categorized as Negative/Dim, Moderate, or Bright for CD19, CD20, and CD22 expression. (K) Schematic illustration of K562 WT cells based on CD20 expression levels, resulting in three populations: CD20 L (low), CD20 M (medium), and CD20 H (high). (L) Violin plots show the percentage of CD20 expression (% CD20 expression) in the sorted K562 WT cell populations, confirming distinct expression levels (n=10). (M) Representative super-resolution microscopy images of differential CD20 surface expression in K562 cells. Images show DAPI (blue, nuclear staining) and CD20 (red) in K562-C20 L (low), K562-C20 M (medium), and K562-C20 H (high) cell. Scale bar indicates 10 μm. (N) Survival of K562 cells expressing varying CD20 expression levels under CAR-T cell treatments. Panels N1 (K562-CD20 L ), N2 (K562-CD20 M ), and N3 (K562-CD20 H ) show the percentage of CD20 + cell survival when treated with Rituximab-based monospecific CAR (Rtx-m20, dark green), in-house humanized anti-CD20 CAR (AB21-m20, green) (N=4). (O) Persistence of CAR-T cells with varying CD20-targeting CAR constructs over 15 days (N=5). Data represents mean ± SEM. ****p < 0.001. A non-parametric t-test was used for statistical analysis between groups. Scale bar indicates 10 μm.

Article Snippet: Briefly, CD19, CD22 CAR expression was evaluated using CD19 and CD20 CAR detection antibodies and CD22 CAR expression (Miltenyi Biotec) was evaluted using Protein L-APC (Cell signaling) followed by PE-conjugated anti-biotin secondary antibodies (Miltenyi Biotec).

Techniques: Expressing, Flow Cytometry, Activity Assay, shRNA, Knockdown, Flux Assay, Phospho-proteomics, Marker, Super-Resolution Microscopy, Staining, Construct

Journal: bioRxiv

Article Title: AI-Guided CAR Designs and AKT3 Degradation Synergize to Enhance Bispecific and Trispecific CAR-T Cell Persistence and Overcome Antigen Escape

doi: 10.1101/2025.06.12.658477

Figure Lengend Snippet: (A) Schematic of the engineering strategy for trispecific CAR-T cells, integrating b20/19-AKT3 PROTAC with a secretory BiTE module consisting of nanobodies targeting CD3 and CD22 (nbCD3/22). (B) Correlation of expression of nbCD3, nb22, CD19 CAR, and CD20 CAR at various MOIs. The cells were treated with Brefeldin and data was obtained using intracellular flow cytometry. (C) Experimental setup for T cell activation, using Jurkat-GFP cells and Dynabeads (db) coated with CD3 to assess secreted nbCD3/22 functionality via flow cytometry. (D) Dose-dependent T cell activation (CD69 expression) in response to culture supernatants with nbCD3/22, using db coated with CD3 for validation. (E) HEK-293T synNotch reporter assay shows dose-dependent inhibition of CD22-CAR signaling by nbCD22 in CAR-T cell supernatants, confirming BiTE functionality under two conditions. (F) Experimental timeline for in vivo CAR-T cell therapy study in Raji WT or NALM6 WT model followed by CAR-T cell administration and TR with Raji CD19/CD20-/- or NALM6 CD19/CD20-/- cells (G) Bioluminescence imaging of Raji and NALM6 tumor-bearing mice treated with b20/19-AKT3 PROTAC or b20/19-AKT3 PROTAC+nbCD3/22 CAR-T cells, monitored from Day 7 to Day 84. (H) Quantified tumor radiance over time, showing sustained tumor control in Raji and NALM6 models with b20/19-AKT3 PROTAC+nbCD3/22 . (I1) Percentage of CAR-T cells in the blood of Raji and NALM6 tumor-bearing mice treated with b20/19-AKT3 PROTAC or b20/19-AKT3 PROTAC+nbCD3/22 , measured over 56 days (I2) Bar graph of CAR-T cell populations in blood at various time points. (J) Levels of nbCD3/22 (pg/mL) in the blood of Raji and NALM6 tumor-bearing mice treated with b20/19-AKT3 PROTAC+nbCD3/22 , measured over 56 days, showing sustained secretion. (K) Kaplan-Meier survival curves demonstrating improved survival with nbCD3/22-modified CAR-T cells. (L) Bar graph and pie charts compare b20/19-AKT3 PROTAC and b20/19-AKT3 PROTAC+nbCD3/22 , showing various memory T cell subsets over time (n=5) in all conditions. Data represents mean ± SEM. ****p < 0.001. A non-parametric t-test was used for statistical analysis between groups.

Article Snippet: Briefly, CD19, CD22 CAR expression was evaluated using CD19 and CD20 CAR detection antibodies and CD22 CAR expression (Miltenyi Biotec) was evaluted using Protein L-APC (Cell signaling) followed by PE-conjugated anti-biotin secondary antibodies (Miltenyi Biotec).

Techniques: Expressing, Flow Cytometry, Activation Assay, Biomarker Discovery, Reporter Assay, Inhibition, In Vivo, Imaging, Control, Modification

Journal: Antibody Therapeutics

Article Title: A bioluminescent reporter bioassay for in-process assessment of chimeric antigen receptor lentiviral vector potency

doi: 10.1093/abt/tbae032

Figure Lengend Snippet: NFAT-Luc2 activity stimulated by multiple T-cell activation signals. (a) Jurkat/NFAT-Luc2 cells were incubated with α-CD3 antibody crosslinked with goat α-mouse IgG antibody at the indicated concentrations. (b) Jurkat/NFAT-Luc2 cells were incubated with or without Raji target cells and the indicated concentrations of blinatumomab. The curves were compared using an extra-sum-of-squares F test and were determined to be different ( P > .0001) (c) Jurkat/NFAT-Luc2 cells were transiently transfected with varying concentrations of α-CD19-CAR or α-CD20-CAR plasmid DNA, balanced with carrier DNA (pGEM-3z) such that all conditions received the same total amount of DNA. One day after transfection, cells were incubated with or without Raji target cells. In all panels, assays were incubated for 6 h at 37°C, and luciferase activity was detected using the Bio-Glo™ Luciferase Assay System. Data are representative of at least three independent experiments with n = 3 technical replicates. Error bars indicate standard deviation. For (c), a one-way ANOVA was performed separately for a-CD19 and a-CD20 conditions. Dunnett’s multiple comparison post-testing found a significant difference between the carrier DNA only (0:1) condition and the CAR DNA conditions in the presence of Raji cells ( P < .0001) and no differences in the absence of Raji cells. Post-testing for linear trend found a relationship between CAR DNA quantity and assay response in the presence of Raji cells ( P < .0001).

Article Snippet: The NFAT-Luc2 reporter plasmid and CD19 and CD20 CAR expression plasmids were generated at Promega Corp. LV particles were generated by VectorBuilder.

Techniques: Activity Assay, Activation Assay, Incubation, Transfection, Plasmid Preparation, Luciferase, Standard Deviation, Comparison