Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
[Preprint]. 2024 Mar 26:2023.12.13.571595.
doi: 10.1101/2023.12.13.571595.

Engineering self-propelled tumor-infiltrating CAR T cells using synthetic velocity receptors

Adrian C Johnston  1 , Gretchen M Alicea  2 , Cameron C Lee  3 , Payal V Patel  1 , Eban A Hanna  1 , Eduarda Vaz  1 , André Forjaz  1 , Zeqi Wan  4 , Praful R Nair  2 , Yeongseo Lim  3 , Tina Chen  1 , Wenxuan Du  1 , Dongjoo Kim  5 , Tushar D Nichakawade  1   6 , Vito W Rebecca  4 , Challice L Bonifant  6 , Rong Fan  5 , Ashley L Kiemen  2   7   6 , Pei-Hsun Wu  1   2 , Denis Wirtz  1   2   7   6
Affiliations

Engineering self-propelled tumor-infiltrating CAR T cells using synthetic velocity receptors

Adrian C Johnston et al. bioRxiv. .

Abstract

Chimeric antigen receptor (CAR) T cells express antigen-specific synthetic receptors, which upon binding to cancer cells, elicit T cell anti-tumor responses. CAR T cell therapy has enjoyed success in the clinic for hematological cancer indications, giving rise to decade-long remissions in some cases. However, CAR T therapy for patients with solid tumors has not seen similar success. Solid tumors constitute 90% of adult human cancers, representing an enormous unmet clinical need. Current approaches do not solve the central problem of limited ability of therapeutic cells to migrate through the stromal matrix. We discover that T cells at low and high density display low- and high-migration phenotypes, respectively. The highly migratory phenotype is mediated by a paracrine pathway from a group of self-produced cytokines that include IL5, TNFα, IFNγ, and IL8. We exploit this finding to "lock-in" a highly migratory phenotype by developing and expressing receptors, which we call velocity receptors (VRs). VRs target these cytokines and signal through these cytokines' cognate receptors to increase T cell motility and infiltrate lung, ovarian, and pancreatic tumors in large numbers and at doses for which control CAR T cells remain confined to the tumor periphery. In contrast to CAR therapy alone, VR-CAR T cells significantly attenuate tumor growth and extend overall survival. This work suggests that approaches to the design of immune cell receptors that focus on migration signaling will help current and future CAR cellular therapies to infiltrate deep into solid tumors.

PubMed Disclaimer

Conflict of interest statement

Conflict of interest statement ACJ and DW are inventors on a patent application based on the work presented here.

Figures

Fig. 1 |
Fig. 1 |. T cells migrate in a density-dependent manner in vitro.
a, Current T cell therapies need to traverse a major physical barrier – the extracellular matrix (ECM) – in the tumor microenvironment (TME) of solid tumors for eventual cell-cell contact that they do not experience in the context of liquid tumors. b, Primary human pan T cells were isolated from primary human PBMCs using a donor leukopak. Isolated T cells premixed with a CD2/CD3/CD28 activation solution and 100 IU/ml IL2 were encapsulated in 3D gels composed of the major ECM protein, type 1 collagen (rat tail type 1 collagen at 2 mg/ml), at increasing T cell densities. c, After a 48-h incubation, the spontaneous motion of T cells at increasing densities (LD = 105 cells/ml, MD = 5×105 cells/ml, HD = 106 cells/ml) embedded in 3D collagen gels was monitored using time-lapsed phase-contrast microcopy for 1 h in 3-min intervals at 37 °C and 5% CO2. Cells were manually tracked to generate trajectories for each cell (see Extended Data Fig. 1). A custom software was used to convert trajectories into mean squared displacement (MSD) for each cell at a time lag of 9 min. Each dot represents an individual cell. LD = low density, MD = medium density, HD = high density. d, A Matlab script was coded to extract the fractions of T cells moving more than their own size (>R2 = 25 μm2, where R = 5 μm is the average T cell radius) from the MSDs shown in panel (c) (see also Extended Data Fig. 1). e, Activated LD and HD T cells in 3D collagen gels were incubated for 48 h, after which medium from LD T cells was replaced with conditioned medium (CM) obtained from HD cells. f, MSDs of LD, HD, and LD + HD CM T cells imaged and analyzed (as in c). g, Motile fractions corresponding to panel (f), computed as in panel (d). For all experiments MSD, median with the SEM are plotted (n=2 technical replicates per biological replicate, N=2 biological replicates). Individual dots represent individual cells; an average of at least 150 cells per technical replicate, corresponding to up to 989 cells per technical replicate were tracked (see source data). One-way ANOVA with Dunn’s multiple comparison test was used for statistical analysis (ns = not significant, *P < 0.01, **P < 0.01, ****P < 0.0001). For all experiments measuring the motile fraction, mean with SEM is plotted (n=2 technical replicates per biological replicate, N=2 biological replicates).
Fig. 2 |
Fig. 2 |. Low and high-motility states of T cell migration in vitro regulated by specific endogenously expressed cytokines.
a, Molecules secreted by LD and HD T cell in 3D collagen gels incubated for 48 h. CM was extracted from the top of LD and HD T cell gels. Single-cell secretomics were run on CM to generate average intensity of a human adaptive immune cytokine panel using Isoplexis Isocode chips. b, Average intensity (from a) normalized per 105 cells for the four cytokines identified (from a) with higher intensities for HD CM compared to LD CM. Data is plotted as mean with SEM (n = 2). One-way ANOVA with Dunn’s multiple comparison test was performed for statistical analysis. ****P < 0.0001. c, Activated LD T cells, from a new donor, in 3D collagen gels were incubated for 48 h. 100 ng/ml IL5, TNFα, IFNγ, or IL8 were then exogenously added to LD T cell gels separately and incubated for 1 h at 37 °C and 5% CO2. d,e, MSDs of T cells in the presence of IL5, TNFα, IFNg, or IL8 (d) and corresponding motile fractions (e). f, Activated HD T cells in 3D collagen gels were incubated for 48 h. 25 mg/ml anti-IL5, anti-TNFαR1, anti-IFNγR1, or anti-IL8R were then exogenously added separately and incubated for 1 h at 37 °C and 5% CO2. g,h, MSDs of HD T cells incubated with different functional antibodies (g) and corresponding motile fractions (h). i, Activated HD T cells, from a new donor, in 3D collagen gels incubated for 48 h. 2.5 mM anti-JAK2 (AG490), 60 mM anti-STAT1 (fludarabine), or 50 mM anti-STAT3 (S3I) were then exogenously added separately and incubated for 1 h at 37 °C and 5% CO2. j,k, MSDs of HD T cell gels incubated with different inhibitors (j) and corresponding motile fractions (k). l, T cells self-propel using velocity cytokines (VCs) IL5, TNFα, IFNγ, and IL8. For all experiments MSD, median with the SEM are plotted (n=2 technical replicates per biological replicate, N=2 biological replicates). Individual dots represent individual cells; an average of at least 80 cells per technical replicate, corresponding to up to 242 cells per technical replicate were tracked (see source data). One-way ANOVA with Dunn’s multiple comparison test was used for statistical analysis (ns = not significant, *P < 0.01, ****P < 0.0001). For all experiments measuring the motile fraction, mean with SEM is plotted (n=2 technical replicates per biological replicate, N=2 biological replicates).
Fig. 3 |
Fig. 3 |. Velocity receptors turn M5CAR T cells into self-propelled highly motile cells and improve their effector functions.
a, Design of velocity receptors (VRs) that bind VCs and signal through VC receptor signaling domains, and development of VR-M5CAR therapies. b, M5CAR or VR-M5CAR T cells, from a new donor, at low density (LD) were encapsulated in 3D collagen gels supplemented with 100 IU/ml IL2 and incubated for 48 h at 37 °C and 5% CO2 and their migration patterns were analyzed as in Fig. 1. c, Control HD M5CAR T cells were also encapsulated in collagen gels and incubated for 48 h and compared to LD VR-M5CAR T cells The spontaneous motion LD M5CAR, LD VR-M5CAR, and HD M5CAR T cells was monitored using time-lapsed phase-contrast microcopy for 1 h at 2-min intervals at 37 °C and 5% CO2. Cells were manually tracked to generate trajectories for each cell (see Extended Data Fig. 3). A custom software was used to convert trajectories into MSD for each cell at a time lag of 10 min. Each dot represents an individual cell. d, Fractions of T cells moving more than their own size (>R2 = 25 μm2, where R = 5 μm is the average T cell radius), as assessed from the MSDs shown in panel (c) (see also Extended Data Fig. 3). e, LD M5CAR cells expressing either VR5αIL5, VR5αTNFα, or VR5αIL8, embedded in 3D collagen gels were incubated for 48 h. 25 μg/ml anti-IL5, anti-TNFα, or anti-IL8 were then exogenously added to VR5αIL5-M5CAR, VR5αTNFα-M5CAR, or VR5αIL8-M5CAR cells, respectively, and incubated for 1 h at 37 °C and 5% CO2. f, MSDs of control LD M5CAR T cells and the different VR-M5CAR T cells shown in (a). T cells were imaged for 30 min at 2-min intervals (as in c). Cell trajectories and MSD were measured and computed (as in c). g, Motile fractions of T cells were obtained (as in d) for data shown in (f). For all experiments, MSD, median with the SEM are plotted (n=2 technical replicates per biological replicate, N=2 biological replicates for (c)). Individual dots represent individual cells; an average of at least 120 cells per technical replicate up to 269 cells per technical replicate were tracked (see source data). One-way ANOVA with Dunn’s multiple comparison test was used for statistical analysis (ns = not significant, *P < 0.05, **P < 0.01, ****P < 0.0001). For experiments measuring the motile fraction, mean with SEM is plotted (n=2 technical replicates per biological replicate, N=2 biological replicates) for (d), and mean is plotted (n=2 technical replicates) for (g). h, Surface expression of the activation markers CD25 and CD69, as determined by flow cytometry using PE anti-CD25 and BV421 anti-CD69. Non-transduced (NTD) T cells were included as a negative control. Surface expression of CD25 was analyzed in FlowJo on M5CAR+-gated T cells as determined by GFP expression. i, Surface expression of the activation maker CD69 was determined by flow cytometry using BV421 anti-CD69. Flow cytometric analysis was performed (as in h). j, In vitro T cell killing of mesothelin-expressing H226 and OVCAR3 cells at the indicated effector-to-target (E:T) ratios. Data plotted as mean ± SEM, n = 3. A two-tailed student’s t test was used for statistical analysis (*P < 0.05, **P < 0.01). k, Il2, TNFα, and IFNγ secretion by non-transduced (NTD), M5CAR T cells, and VR-M5CAR T cells after overnight co-incubation with mesothelin-expressing ASPC1, H226, and OVCAR3 cells. Cytokine release was measured using ELISA. Mean with SEM is shown (n = 3). One-way ANOVA with Tukey’s multiple comparison test was used for statistical analysis (ns = not significant, *P < 0.05, **P < 0.01, ***P<0.001, ****P < 0.0001).
Fig. 4 |
Fig. 4 |. VRs greatly elevate M5CAR T cell tumor infiltration for improved controlled tumor growth in a pancreatic cancer model.
a, 2×106 ASPC1 pancreatic cancer tumors pre-mixed in 1:1 Matrigel:PBS were subcutaneously (s.c.) engrafted into 8–12 week old NSG mice. Tumor volumes were measured twice a week using digital calipers. When tumors were palpable (100 – 250 mm3), mice were randomized and blindly treated 10 days later with a single intravenous (i.v.) dose of 3×106 NTD or M5CAR T cells. 25 days after tumor engraftment, tumors were harvested and sent for sectioning and IHC staining for human CD45 at the Johns Hopkins Oncology Tissue Services Core Facility. IHC-stained tissue slides were scanned, and images were extracted using ImageScope (middle panel). Tumor volumes were calculated as L×W2×0.5 plotted as mm3 (right panel). b, ASPC1-bearing mice were treated with M5CAR alone or VR5αTNFα–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). c, ASPC1-bearing mice were treated with M5CAR alone or VR5αIL8–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). d, ASPC1-bearing mice were treated with M5CAR T cells alone or Vγsig–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). e, ASPC1-bearing mice were treated with M5CAR alone or VR5αIL5–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). f, ASPC1-bearing mice were treated with M5CAR alone or V5–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). g, Human CD45+ cell numbers were obtained as described in materials and methods. Tissue area was computationally obtained. Tumor infiltration of human CD45+ cells per area was calculated and plotted. h, Human CD45+ cells per tumor area were plotted against average tumor volume at the study endpoint. For figure panels a-f, IHC scale bars are 2 mm (left) and 100 μm (right). Tumor volumes are plotted as mean ± SEM (n = 5 mice per group, except for NTD for which n = 4). Two-tailed student’s t test was used for statistical analysis (ns = not significant, *P < 0.05, **P < 0.01, ***P <0.001). For calculation of cells/cm2 in figure panels g,h, n = 6 for NTD or n = 8 for VR-M5CARs, with 2 non-consecutive tissue slides stained per mouse and 3–4 mice per group; n = 7 for M5CAR alone, with 2 non-consecutive tissue slides stained for 3 mice and 1 tissue slide stained for 1 mouse. See Extended Data Fig. 5a for individual tumor growth curves.
Fig. 5 |
Fig. 5 |. VRs improve M5CAR T cell tumor infiltration which slows tumor growth in a lung cancer model.
a, 5×106 H226 lung cancer tumors pre-mixed in 1:1 Matrigel:PBS were engrafted s.c. into 8–12 week old NSG mice. Tumor volumes were measured twice a week using digital calipers. When tumors were palpable (100 – 250 mm3), mice were randomized and blindly treated (9 days later) with a single i.v. dose of 3×106 NTD or M5CAR T cells. 19 days after tumor engraftment, tumors were harvested and sent for sectioning and IHC staining for human CD45 at the Johns Hopkins Oncology Tissue Services Core Facility. IHC-stained tissue slides were scanned, and images were extracted using ImageScope (middle panel). Tumor volumes were calculated as L×W2×0.5 plotted as mm3 (right panel). b, H226-bearing mice were treated with M5CAR alone or with VR5α TNFα–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained as in panel A. c, Mice bearing H226 tumors were treated with M5CAR alone or Vγsig–M5CAR T cells (as in panel a). IHC images and tumor volumes were obtained (as in panel a). d, Mice bearing H226 tumors were treated with M5CAR alone or VR5α IL8–M5CAR T cells as in panel A. IHC images and tumor volumes were obtained (as in panel a). e, H226-bearing mice were treated with M5CAR alone or V5–M5CAR T cells as in panel A. IHC images and tumor volumes were obtained (as in panel a). f, Mice bearing H226 tumors were treated with M5CAR alone or VR5α IL5–M5CAR T cells as in panel A. IHC images and tumor volumes were obtained (as in panel a). g, Human CD45+ cell numbers were obtained as described in materials and methods. Tissue area was computationally obtained. Tumor infiltration of human CD45+ cells per area was calculated and plotted. h, Human CD45+ cells per tumor area were plotted against average tumor volume at the study endpoint. For figure panels a-f, IHC panel scale bars are 2 mm (left) and 100 μm (right). Tumor volumes are plotted as mean ± SEM (n = 5 mice per group, except for NTD for which n = 4). Two-tailed student’s t test was used for statistical analysis (ns = not significant, *P < 0.05, **P < 0.01). For calculation of cells/cm2 in figure panels g,h, n = 6 for NTD or n = 8 for M5CAR alone and VR-M5CARs, with 2 non-consecutive tissue slides stained per mouse and 3–4 mice per group. See Extended Data Fig. 5b for individual tumor growth curves.
Fig. 6 |
Fig. 6 |. VRs-M5CAR T cells extend the overall survival of mice in an ovarian cancer model.
a, 5×106 OVCAR3 ovarian cancer tumor cells pre-mixed in 1:1 Matrigel:PBS were engrafted s.c. into 8–12 week old female NSG mice. Tumor volumes were measured twice a week using digital calipers. When tumors were palpable (100 – 250 mm3), mice were randomized and blindly treated (8 days later) with a single i.v. dose of 3×106 M5CAR alone or VR-M5CAR T cells. Mice were treated 19 days later with 3×106 M5CAR alone or VR-M5CAR T cells by intraperitoneal (i.p.) injection. Mice reached an endpoint when OVCAR3 tumors reached 750 mm3 in size. b, OVCAR3-bearing mice were treated with M5CAR alone or VR5α TNFα–M5CAR T cells. Tumor volumes were obtained as described in (a). Mice that reached the endpoint tumor size as described in panel (a) were removed from the study. Plotted are Kaplan-Meier survival curves. c, OVCAR3-bearing mice were treated with M5CAR alone or Vγsig–M5CAR T cells. Kaplan-Meier survival curves were generated as described in panel (b). d, Mice bearing OVCAR3 tumors were treated with M5CAR alone or V5-M5CAR T cells. Kaplan-Meier survival curves were generated as described in panel (b). e, OVCAR3-bearing mice were treated with M5CAR alone or VR5α IL8M5CAR T cells. Kaplan-Meier survival curves were generated as described in panel (b). f, Mice bearing OVCAR3 tumors were treated with M5CAR alone or VR5α IL5-M5CAR T cells Kaplan-Meier survival curves were generated as described in panel (b). All plots show Kaplan-Meier survival curves (n = 5 mice per group). Log-rank Mantel-Cox test was used for statistical analysis (ns = not significant, **P < 0.01)
See this image and copyright information in PMC

Similar articles

See all similar articles

References

    1. June C. H. & Sadelain M. Chimeric Antigen Receptor Therapy. N Engl J Med 379, 64–73 (2018). 10.1056/NEJMra1706169 - DOI - PMC - PubMed
    1. Melenhorst J. J. et al. Decade-long leukaemia remissions with persistence of CD4. Nature 602, 503–509 (2022). 10.1038/s41586-021-04390-6 - DOI - PMC - PubMed
    1. Sterner R. C. & Sterner R. M. CAR-T cell therapy: current limitations and potential strategies. Blood Cancer J 11, 69 (2021). 10.1038/s41408-021-00459-7 - DOI - PMC - PubMed
    1. Siegel R. L., Miller K. D., Wagle N. S. & Jemal A. Cancer statistics, 2023. CA Cancer J Clin 73, 17–48 (2023). 10.3322/caac.21763 - DOI - PubMed
    1. Haas A. R. et al. Two cases of severe pulmonary toxicity from highly active mesothelin-directed CAR T cells. Mol Ther 31, 2309–2325 (2023). 10.1016/j.ymthe.2023.06.006 - DOI - PMC - PubMed

Publication types