Journal: bioRxiv

Article Title: A scalable tumor-vasculature-on-chip for CAR T cell trafficking and efficacy studies

doi: 10.64898/2026.02.05.703975

Figure Lengend Snippet: (A) Schematic representation of the concept of immune checkpoint inhibition. Interaction between PD-1 and PD-L1 is blocked by Nivolumab which restores release of cytotoxic molecules by the CAR T cell (red) (B) Heatmap showing an overview of changes in tumor area and release of Granzyme B, IL-6, sPD-L1 and TNFα in HT-29-GFP and HCT116-GFP co-cultures in response to both single and combination treatments with Nivolumab, Ipilimumab and Temozolomide. Data used to generate the heatmap can be found in supplementary figures 7 and 8. (C,D) PCA plots showing response of HT-29-GFP and HCT116-GFP co-cultures to EpCAM-CD28 CAR T cells and treatments. (E) PCA plot showing the response of HT-29-GFP and HCT116-GFP co-cultures to combination treatments. Datapoints are scaled for the initial tumor cell density at the start of the co-culture. Pink circle indicates a small cluster of replicates consisting of HCT116-GFP cultures exposed to Nivolumab, Ipilimumab and Temozolomide. (F) Contribution and directionality of the different features of the principal components. (G) Heatmap showing morphometric parameters measured in the endothelial vessels of HT-29-GFP and HCT116-GFP co-cultures exposed to EpCAM-CD28 CART cells and treatments. Shown are fold changes against untreated controls. (H) Correlation distance matrix showing similarity between treatment responses based on patterns in relative change in tumor area, cytokine release and endothelial response. Data are presented as unsquared correlation distances, with values <1 indicating similarity in response while values >1 indicate distinct responses. A value of 1 means no relation between observed responses and a value of 0 indicates an identical response.

Article Snippet: Human colorectal cancer cell line HCT116-GFP (GeneCopoeia, SL024) was cultured in RPMI-1640 (Gibco, 11875093) supplemented with 10% Fetal Bovine Serum and 1 % Penicillin-Streptomycin.

Techniques: Inhibition, Co-Culture Assay

Journal: bioRxiv

Article Title: A scalable tumor-vasculature-on-chip for CAR T cell trafficking and efficacy studies

doi: 10.64898/2026.02.05.703975

Figure Lengend Snippet: (A) HCT116-GFP tumor cells proliferate after introduction in the lumenised ECM and cause regression of the endothelial vessel over time. The fluorescent signal could be used to create a binary mask that was used to track growth of HCT116-GFP tumor cells from day 5 until day 8. Fluorescent area at each timepoint was normalized against the area at t=1 for each individual replicate (n=15). (B) Effect of treatments on HT-29-GFP tumor area. Data was normalized against the untreated control at each timepoint (n=8-16). (C) Effect of treatments on HCT116-GFP tumor area. Data was normalized against the untreated control at each timepoint (n=8-14). Data included in the graphs are presented as mean ± SD.

Article Snippet: Human colorectal cancer cell line HCT116-GFP (GeneCopoeia, SL024) was cultured in RPMI-1640 (Gibco, 11875093) supplemented with 10% Fetal Bovine Serum and 1 % Penicillin-Streptomycin.

Techniques: Control

Journal: bioRxiv

Article Title: A scalable tumor-vasculature-on-chip for CAR T cell trafficking and efficacy studies

doi: 10.64898/2026.02.05.703975

Figure Lengend Snippet: (A) Analysis of cytokine levels (pg/mL) in media samples after 72 hours of co-culture of HT-29-GFP and HCT116-GFP tumor cells with EpCAM-CD28 CAR T cells. Co-cultures were untreated or exposed to Nivolumab, Ipilimumab and/or Temozolomide (n=8-14). (B) Cytokine levels in panel A were normalized against untreated controls per experiment resulting in fold changes that were used for principal component analysis (n=8-14). Data included in the graphs are presented as mean ± SD.

Article Snippet: Human colorectal cancer cell line HCT116-GFP (GeneCopoeia, SL024) was cultured in RPMI-1640 (Gibco, 11875093) supplemented with 10% Fetal Bovine Serum and 1 % Penicillin-Streptomycin.

Techniques: Co-Culture Assay

Journal: bioRxiv

Article Title: A scalable tumor-vasculature-on-chip for CAR T cell trafficking and efficacy studies

doi: 10.64898/2026.02.05.703975

Figure Lengend Snippet: (A) Endothelial vessels were co-cultured with either HT-29-GFP or HCT116-GFP tumor cells and stained with VE-cadherin after the 72-hour co-culture. Co-cultures were either untreated or exposed to single or combination treatments with Nivolumab, Ipilimumab and Temozolomide in the presence of EpCAM-CD28 CAR T cells (E:T = 1:1). The endothelial response was assessed by analysis of VE-cadherin objects using a range of different morphological and spatial descriptors using IN Carta® (n=7-14) and normalization against respective untreated controls containing CAR T cells for each parameter. Data are presented as mean ± SD. (B) Heatmap showing change in morphometric parameters measured in endothelial vessels, co-cultured with either HT-29-GFP or HCT116-GFP tumor cells, upon addition of EpCAM-CD28 CAR T cells (E:T = 1:1). Data is presented as fold changes against respective controls without CAR T cells.

Article Snippet: Human colorectal cancer cell line HCT116-GFP (GeneCopoeia, SL024) was cultured in RPMI-1640 (Gibco, 11875093) supplemented with 10% Fetal Bovine Serum and 1 % Penicillin-Streptomycin.

Techniques: Cell Culture, Staining, Co-Culture Assay

Journal: Cell reports

Article Title: Proteasome activity maintains cell-type-specific gene expression

doi: 10.1016/j.celrep.2026.116973

Figure Lengend Snippet: (A) Lysates from HCT116 WT, ΔRAZUL, or hRpn10 VWA cells were resolved and analyzed by immunoprobing for hRpn10, E6AP, or β-actin. (B) Lysates from WT, ΔhRpn13, trRpn13, ΔRAZUL, hRpn10 VWA , or ΔUCHL5 cells were resolved and immunoprobed for K48-linked (left) or bulk (right) ubiquitin chains or β-actin. (C) Immunoblots probing for ubiquitin or hRpn2 of proteasomes immunoprecipitated by anti-hRpt3 antibodies from WT, ΔhRpn13, trRpn13, ΔRAZUL, hRpn10 VWA , or ΔUCHL5 without (left) or with (right) DSP crosslinker. IgG antibodies were included as a control. (D) Boxplot representation plotting the range and mean value of protein abundance for hRpn1, hRpn10, hRpn8, hRpn11, hRpn13, hRpn2, hRpt3, β5, E6AP, UCHL5, and USP14 across three samples as measured by TMT-MS. (E) Lysates from HCT116 WT, ΔhRpn13, trRpn13, ΔRAZUL, hRpn10 VWA , or ΔUCHL5 cells resolved and analyzed by immunoprobing for hRpn1, hRpn2, hRpn10, hRpn13, hRpn8, hRpn11, USP14, UCHL5, β5, or β-actin. (F) Graphical plot of protein and mRNA abundance in hRpn10 VWA compared to WT by TMT-MS ( x axis) and qPCR ( y axis), respectively, for RP subunits hRpn1, hRpn10, hRpn11, and hRpn13, CP β5, or deubiquitinases UCHL5 and USP14. A dotted line indicates no change. (G) Volcano plot of fold-change in protein abundance ( x axis, log 2 ) and adjusted p value ( y axis, −log 10 ) in hRpn10 VWA cells relative to WT. Vertical and horizontal lines are included at ±1 and 1.3, respectively. Proteins with expression values statistically greater or less than 2-fold difference are colored and labeled according to their molecular function as indicated in the legend. (H) Plot of the change in protein abundance (log 2 ) relative to WT by TMT-MS for ΔRAZUL ( x axis) and hRpn10 VWA ( y axis). The Spearman rank correlation value is included. Data with adjusted p value ≥ 0.05 are excluded from this analysis. Color coding follows (G). FC, fold change. (I) Illustration of the aftereffects that arise following deletion of hRpn10 UIMs and RAZUL domain. Weak binding of ubiquitinated substrates to proteasomes causes dysregulated protein levels in multiple functional pathways and induces upregulated gene expression of proteasome subunits. hRpn10 VWA , ubiquitin, substrate, peptide, RP, and CP α or β subunits are colored purple, yellow, brown, dark gray, light yellow, gray, or light gray, respectively. See also , , and .

Article Snippet: Cultured human HCT116 colorectal carcinoma line Libraries were produced by the AZENTA/GENEWIZ using a Nextera/Proprietary method.

Techniques: Ubiquitin Proteomics, Western Blot, Immunoprecipitation, Control, Quantitative Proteomics, Expressing, Labeling, Binding Assay, Functional Assay, Gene Expression

Journal: Cell reports

Article Title: Proteasome activity maintains cell-type-specific gene expression

doi: 10.1016/j.celrep.2026.116973

Figure Lengend Snippet: (A) Replot of , proteins with statistically different changes in protein expression are colored by whether they are tissue-specific (red), not tissue-specific (blue), or not present in the RNA-seq database used for tissue-specificity determination (NA, not available, black). (B) Heatmap of the normalized transcript per million (nTPM) values across 50 different tissue types for the most significantly affected proteins (|log 2 | ≥ 2) in (A). For each protein, nTPM values were normalized to the highest nTPM value across all tissues. Color scale indicates the normalized nTPM values, with white representing lower and red representing higher values as indicated. (C–E) Immunoblots of lysate from HCT116 WT and hRpn10 VWA detecting PALM3, SUSD2, CDK6, SEC31A, RAB25, S100A14, hRpn10, and β-actin. (F) Model of proteasome activity contributing to cell identity that is represented by colored spots inside squares. A cartoon of a cell with the cytosol (yellow) and nucleus (light yellow) is displayed and an aberrant proteome (represented by little color variance) for cells with hRpn10 VWA proteasomes contrasted with a healthy proteome for cells with WT proteasomes. See also and .

Article Snippet: Cultured human HCT116 colorectal carcinoma line Libraries were produced by the AZENTA/GENEWIZ using a Nextera/Proprietary method.

Techniques: Expressing, RNA Sequencing, Western Blot, Activity Assay

Journal: Cell reports

Article Title: Proteasome activity maintains cell-type-specific gene expression

doi: 10.1016/j.celrep.2026.116973

Figure Lengend Snippet: (A) Lysates from HCT116 WT or hRpn10 VWA cells treated with 100 nM carfilzomib (CZ) for 24 h were immunoprobed for SUSD2, PALM3, hRpn10, or β-actin. (B) Bar plot of mRNA abundance for the indicated genes in HCT116 WT (black) and hRpn10 VWA (pink) cells. The data represent mean ± SD of n = 3 biological replicates. **** p value < 0.0001; two-tailed unpaired t test in GraphPad Prism 10. (C) Graphical plot of mRNA (log 2 FC, x axis) and protein abundance (log 2 FC, y axis) in hRpn10 VWA compared to WT cells by qPCR and TMT-MS, respectively, for the indicated genes, colors as in . (D) Bar plot of mRNA abundance for SUSD2 and PALM3 in HCT116 WT cells treated with DMSO (black) or 100 nM carfilzomib (CZ) (gray) for 24 h. The data represent mean ± SD of n = 3 biological replicates, **** p value < 0.0001 and ** p value = 0.0043; two-tailed unpaired t test in GraphPad Prism 10. (E) Bar plot of mRNA abundance for SUSD2 , PALM3 , VSNL1 , and NEBL in HEK293T WT (black) and HEK293T VWA/WT (pink) cells treated with 50 nM scramble or hRpn10 siRNA for 48 h, as indicated. The data represent mean ± SD of n = 3 biological replicates. **** p value < 0.0001, *** p value = 0.0006, ** p value < 0.005, and * p value = 0.0218; ns, not significant; one-way ANOVA for multiple comparisons in GraphPad Prism 10. (F) Lysates from HEK293T WT or HEK293T VWA/WT cells transfected with 50 nM hRpn10 or scramble (control) siRNAs for 48 h were immunoprobed for PALM3, hRpn10, SUSD2, OTUD5, or β-actin. #1 and #2 represent SUSD2 fragments. (G) Model of proteasomes with defective hRpn10 causing failed protein turnover and, in turn, transcriptional dysregulation and loss of a cell-type-specific proteome. Colored follow and . See also and .

Article Snippet: Cultured human HCT116 colorectal carcinoma line Libraries were produced by the AZENTA/GENEWIZ using a Nextera/Proprietary method.

Techniques: Two Tailed Test, Quantitative Proteomics, Transfection, Control

Journal: Cell reports

Article Title: Proteasome activity maintains cell-type-specific gene expression

doi: 10.1016/j.celrep.2026.116973

Figure Lengend Snippet: (A) Lysates from WT, ΔhRpn13, trRpn13, ΔRAZUL, hRpn10 VWA , or ΔUCHL5 were resolved and analyzed by immunoprobing for OTUD5, hRpn13, UCHL5, or β-actin. (B) Immunofluorescence of HCT116 WT and hRpn10 VWA cells, immunostained for OTUD5 (red in composite), α-tubulin (white in composite), and counterstained with DAPI (blue in composite). Individual gray-scale images are shown for each channel, followed by composites of OTUD5 and DAPI and a complete composite image of all channels. Scale bars (bottom left in WT DAPI and bottom right in hRpn10 VWA DAPI), 10 μm. (C) Box and whiskers plot of correlation coefficient of OTUD5 and DAPI in HCT116 WT and hRpn10 VWA cells. The data represent min to max of n = 3 biological replicates, * p value = 0.024; two-tailed unpaired t test in GraphPad Prism 10. (D) Re-probing the membrane in (left) for OTUD5 and β-actin. (E) Bar plot of mRNA abundance for OTUD5 in HCT116 WT (black) and hRpn10 VWA (pink) cells. The data represent mean ± SD of n = 6 biological replicates; ns, not significant; two-tailed unpaired t test in GraphPad Prism 10. (F) Lysates from HCT116 WT or hRpn10 VWA cells transfected with 50 nM OTUD5 or scramble (control) siRNAs for 48 h were immunoprobed for SUSD2 (short and long exposure), OTUD5, PALM3, hRpn10, or β-actin. (G) Bar plot of mRNA abundance for SUSD2, PALM3 , and VSNL1 in HCT116 WT (black) and hRpn10 VWA (pink) cells treated with 50 nM scramble or OTUD5 siRNAs for 48 h. The data represent mean ± SD of n = 3 biological replicates, **** p value < 0.0001; ns, not significant; one-way ANOVA for multiple comparisons in GraphPad Prism 10. (H) Illustration of hRpn10 VWA proteasomes having impaired OTUD5 turnover and, in turn, dysregulated transcription for a subset (indicated by a thinner arrow) of proteins. Colored as in . See also .

Article Snippet: Cultured human HCT116 colorectal carcinoma line Libraries were produced by the AZENTA/GENEWIZ using a Nextera/Proprietary method.

Techniques: Immunofluorescence, Two Tailed Test, Membrane, Transfection, Control

Journal: Cell reports

Article Title: Proteasome activity maintains cell-type-specific gene expression

doi: 10.1016/j.celrep.2026.116973

Figure Lengend Snippet: (A) Graphical plot of protein abundance change comparing hRpn10 VWA to WT filtering by adjusted p value < 0.05 for 743 proteins commonly detected by DIA-MS (log 2 FC, x axis) and TMT-MS (log 2 FC, y axis). The DIA-MS data were taken from hRpn10 VWA and WT cells treated with scramble control siRNAs. Protein labels are colored according to tissue specificity, red indicating selective expression by tissue type, and blue for broad expression. NA, not applicable (black). The Spearman coefficient is included. (B) Graphical plot of DIA-MS data for 48 proteins for which protein levels are restored by OTUD5 suppression, showing protein abundance change (log 2 ) comparing hRpn10 VWA to WT in siRNA scramble-treated cells ( x axis) and hRpn10 VWA with ( OTUD5 siRNA treatment) and without (siRNA scramble) OTUD5 suppression ( y axis). Coloring follows (A). The Spearman coefficient is included. (C) Heatmap of the data plotted in (B), separated as 26 downregulated (left) and 22 upregulated (right) proteins in hRpn10 VWA cells. A scale bar is included below, and tissue-specific proteins are labeled in red. (D) Heatmap showing the relative expression of OTUD5, known OTUD5 chromatin regulator substrates modified by K48 or K63-linked ubiquitin chains, selected OTUD5-dependent or independent proteins as well as the indicated controls. (E and F) Lysates from HCT116 WT or hRpn10 VWA cells transfected with 50 nM OTUD5 (E), UBR5 (F), HDAC2 (F), or scramble (control) siRNAs for 48 h were immunoprobed for PALM3, HDAC2, OTUD5, SUSD2 (short and long exposure), UBR5, or β-actin. (G) Bar plot of mRNA abundance for SUSD2 , PALM3 , and VSNL1 in HCT116 WT (black) and hRpn10 VWA (pink) cells treated with 50 nM scramble, UBR5 , or HDAC2 siRNAs for 48 h. The data represent mean ± SD of n = 3 biological replicates. **** p value < 0.0001, ** p value = 0.0022, and * p value = 0.03; ns, not significant; one-way ANOVA for multiple comparisons in GraphPad Prism 10. (H) Schematic showing OTUD5 (orange) regulating HDAC2 (olive) and UBR5 (blue) by deubiquitination on chromatin (black), thereby influencing transcription of genes canonically expressed with cell type specificity. UBR5 reciprocally ubiquitinates OTUD5, causing its degradation. Data in (C and D) are expressed as the percentage of deviation when compared against a baseline averaged across all samples. The color scale indicates the percentage of deviation from the baseline, with blue representing lower and red representing higher values as indicated. See also and .

Article Snippet: Cultured human HCT116 colorectal carcinoma line Libraries were produced by the AZENTA/GENEWIZ using a Nextera/Proprietary method.

Techniques: Quantitative Proteomics, Control, Expressing, Labeling, Modification, Ubiquitin Proteomics, Transfection