Cell expansion platform

Abstract

A hydrogel comprising (a) polyethylene glycol diacrylate (PEGDA), polyethylene monoacryate (PEGMA), and / or Poly(N-isopropylacrylamide) (PNIPAM), (b) streptavidin acrylamide, and (c) a photoinitiator. In certain aspects, the composition comprising a solid substrate coated with the hydrogel. In certain aspects, provided herein is a method of making a functionalized hydrogel thin film, and a functionalized hydrogel thin film made by this method. In certain aspects, provided herein is a method of expanding T cells comprising, seeding cells onto the functionalized hydrogel thin film substrates.

Description

[0001] CELL EXPANSION PLATFORM

[0002] CROSS-REFERENCE TO RELATED APPLICATION

[0003] This application claims priority to United States Provisional Application Number 63 / 779,041 which was filed on March 27, 2025. The entire content of the application referenced above is hereby incorporated by reference herein.

[0004] BACKGROUND

[0005] Systems for activating and expanding cell populations are useful for several applications. For example, mesenchymal stem cells (MSCs) are useful for tissue engineering, B cells for antibody production, non-mammalian cells for small molecule production and immune cells for re-infusion via adoptive immunotherapy. A current manufacturing bottleneck is the safe and rapid proliferation of cells. Accordingly, new compositions and methods to expand target cell populations are needed.

[0006] SUMMARY

[0007] In certain aspects, provided herein is a hydrogel with orthogonal control of mechanics and ligand presentation comprising (a) polyethylene glycol diacrylate (PEGDA), polyethylene monoacryate (PEGMA), and / or Poly(N-isopropylacrylamide) (PNIPAM), (b) streptavidin acrylamide, and (c) a photoinitiator.

[0008] In certain aspects, provided herein is a composition comprising a solid substrate coated with the hydrogel as described herein.

[0009] In certain aspects, provided herein is a method of making a functionalized hydrogel thin film comprising:

[0010] a. consecutively treating a glass slide with 0.1 M NaOH, 2% 3-aminopropyltrimethoxysilane (APTMS), and 0.5% glutaraldehyde, with washing / drying steps in between to form a silane-functionalized slide,

[0011] b. mixing PEGDA, a photoinitiator, and streptavidin acrylamide in molecular biology grade water to form a hydrogel precursor solution,

[0012] c. placing a 20-200 pl droplet of the hydrogel precursor solution between a first silane-functionalized slide and a second silane-functionalized slide to form a sandwich, d. exposing the sandwich to light,

[0013] e. incubating the sandwich in PBS at 4°C to 20°C for 1- 24 hours,f. removing one of the silane-functionalized slides from the sandwich to expose a thin film gel,

[0014] g. sterilizing the thin film gel,

[0015] h. contacting biotinylated molecules to the thin film gel to form a hydrogel substrate,

[0016] i. incubating the hydrogel substrate were incubated for about 1-16 hours at about 4-37°C with shaking, and

[0017] j . washing the hydrogel substrate with PBS and cell culture media to form a functionalized hydrogel thin film.

[0018] In certain aspects, provided herein is a functionalized hydrogel thin film made by the process described herein.

[0019] In certain aspects, provided herein is a method of expanding T cells comprising, seeding cells onto the functionalized hydrogel thin film substrates described herein.

[0020] BRIEF DESCRIPTION OF DRAWINGS

[0021] Figures 1A-1E. PEGDA-Strep hydrogels are mechanically tunable and functionalizable hydrogels. Fig. 1A. PEGDA-Strep hydrogel substrate design and fabrication. Fig. IB. PEGDA-Strep crosslinking reaction. Fig. 1C. Elastic modulus of PEGDA-Strep scaffolds measured as a function of monomer concentration (Data are mean ± SD for n=4 gels across 4 independent experiments. Fig. ID. Coating profiles of hydrogels across all formulations (Data are mean ± SD for 5 gels across 4 independent experiments). Fig. IE. Fluorescent microscopy analysis of mean fluorescence intensity at hydrogel interface across PEGDA-Strep gel formulations with fluorescent bound ligand at increasing incubation concentrations (Data are mean ± SD for 5 gels across 4 independent experiments, Two-way ANOVA (u = 0.05)).

[0022] Figures 2A-2D. Engagement of adhesion receptors LFA-1 or CD2 mechanosensitively enhances human primary T cell activation. Fig. 2A. Diagram of T cell: dendritic cell immune synapse and PEGDA-Strep scaffold incorporating engagement of CD2 and LFA-1 during T cell activation. Fig. 2B CD69 expression, Fig. 2C CD25 expression, and Fig. 2D IL-2 secretion in T cells incubated for 20 hours coated PEGDA-Strep gels of varying stiffness at lx ligand density. (Data are mean ± SD, n =6 distinct donors, Two-way ANOVA, **p<0.01, ***p<0.005, n.s. not significant, o=0.05, Tukey post hoc comparison test)Figures 3A-3G. Effects of stiffness, adhesion receptor engagement, and ligand density on activation of primary human T cells using PEGDA-Strep gels (Fig. 3A) Diagram of ligand density and substrate stiffness titration early activation experiment (Figs.

[0023] 3B-3D) CD69 expression in T cells incubated for 20 hours coated PEGDA-Strep gels grouped by ligand density as a function of stiffness for (Fig. 3B) anti-CD3 / CD28 alone, (Fig.

[0024] 3C) added LFA-1 engagement, and (Fig. 3D) added CD2 engagement. (Figs. 3E-3G) CD25 expression in T cells incubated for 20 hours coated PEGDA-Strep gels grouped by ligand density in response to increasing stiffness for (Fig. 3E) anti-CD3 / CD28 alone, (Fig. 3F) added LFA-1 engagement, and (Fig. 3G) added CD2 engagement. (Data are mean ± SD, n =6 distinct donors, Two-way ANOVA followed by post-hoc t-test, *p<0.05, **p<0.01, ****p<0.001, n.s. not significant, o=0.05)

[0025] Figures 4A-4I. Engagement of adhesion receptors LFA-1 and CD2 enhances T cell proliferation (Fig. 4A) Timeline of T cell proliferation experiments. (Fig. 4B) Representative brightfield microscopy images of live T cells following 3-day activation on PEGDA-Strep gels, scale bar = 100 pm. (Fig. 4C) Representative confocal fluorescent microscopy images of live T cells stained with Hoechst nuclear stain (blue), F-actin stain (green), and tubulin stain (red) following 3-day activation on PEGDA-Strep gels, scale bar = 40 pm. (Figs. 4D-4F) Assessment of cell proliferation 72-hours post cell seeding on gel. (Fig. 4D) Representative CFSE proliferation histograms following T cell activation on PEGDA-Strep hydrogels of varying stiffness and ligand presentation. (Fig. 4E) Percent dividing cells and (Fig. 4F) CD25 expression following 72-hour CFSE proliferation assay of T cell cultures activated on PEGDA-Strep hydrogels of varying stiffness and ligand presentation (Data are mean ± SD, n = 6 distinct donors, Two-way ANOVA followed by post-hoc t-test, **p<0.01, ***p<0.005, o=0.05). (Figs. 4G-4I) 13-day expansion of primary human T cells activated on PEGDA-Strep gels of varied stiffness and ligand presentation. (Fig. 4G) Representative expansion kinetics diagram showing differences in expansion between stiffness and adhesion receptor engagement conditions. (Fig. 4H) 13-day fold expansion of primary human T cells (Data are mean ± SD, n=6-8 distinct donors, two-way ANOVA followed by post-hoc t-test, *p<0.05, **p<0.01, o=0.05). (Fig. 41) Matched donor comparison of stiff PEGDA-Strep hydrogels with LFA-1 / CD2 engagement to Dynabead fold expansion (n=8 distinct donors, Student’s t-test, *p<0.05, **p<0.01, o=0.05)

[0026] Figures 5A-5J. Assessment of T cell function using a transgenic mouse model(5A) Schematic and timeline of in vitro cell killing assay. (Figs. 5B, 5C) Quantification of in vitro killing of non-pulsed or SIINFEKL-pulsed B16-F10-Luc target cells by CD8+ OT-I T cells activated on PEGDA-Strep gels of (Fig. 5B) 6kPa or (Fig. 5C) 60kPa and then cocultured for 24 hours at an effectortarget cell ratio of 1 : 1. (Data are mean ± SD, n=3 mice) (Figs. 5D-5G) Schematic and timeline of long term expansion and in vitro cell killing experiments with OT-1 mouse cells. (Fig. 5E) 11-day fold expansion of primary mouse OT-I CD8+ T cells activated on PEGDA-Strep gels of varied stiffness and ligand presentation. (Fig. 5F) Quantification of in vitro killing of SIINFEKL-pulsed Bl 6-F10-Luc target cells by CD8+ OT-I T cells expanded for 11 days following activation on PEGDA-Strep gels and then co-cultured at an effector Target cell ratio of 10:1. (Fig. 5G) Quantification of in vitro killing of SIINFEKL-pulsed Bl 6-F10-Luc target cells by CD8+ OT-I T cells following activation at day 1 or expanded for 11 days following activation on PEGDA-Strep gels and then co-cultured at an effectortarget cell ratio of 5:1. (Figs. 5H-5J) Quantification of in vitro killing of SIINFEKL-pulsed or unpulsed Bl 6-F10-Luc target cells by CD8+ OT-I T cells following activation on PEGDA-Strep gels, expansion for 11 days, and co-cultured at an effector target cell ratio of (Fig. 5H) 5:1, (Fig. 51) 1:1 or (Fig. 5 J) 1:10. (Data are mean ± SD, n=3-4 mice, Two-way ANOVA followed by post-hoc t-test, *p<0.05, **p<0.01, ***p<0.005, a=0.05).

[0027] Figures 6A-6H. Phenotypic characterization of T cell cultures expanded on PEGDA-Strep hydrogels (Fig. 6A) T cell expansion phenotype characterization experiment diagram. (Fig. 6B) Ratios of CD4+:CD8+ T cells in 13-day expanded T cell cultures across varied ligand presentation conditions and Dynabead controls. (Fig. 6C) Comparison of proportions of CD4+ TH2-like CXCR3-CCR4+CCR6+ phenotypes. (Fig. 6D) Comparison of proportions of T cells expressing IFN-y across conditions. (Fig. 6E) Characterization of Tscm (CCR7+CD45RA+), Tern (CCR7+CD45RA-), Tern (CCR7-CD45RA-), and Temra(CCR7-CD45RA+) subset proportions in 13-day expanded T cells across varied ligand presentation conditions and Dynabead controls. (Fig. 6F) Comparison of Tscm subset proportions across conditions. (Fig. 6G) Comparison of proportions of CD 127+ T cells across conditions. (Fig. 6H) Comparison of proportions of CD27+ T cells across conditions. Data in are mean ± SD, n =4 distinct donors Data were analyzed using Two-way ANOVA followed by post-hoc t-test, *p<0.05, **p<0.01, ***p<0.005, n.s. not significant, o=0.05).

[0028] Figure 7. FoxP3+ regulatory T cells can be enriched during expansion from primary blood mononuclear cells.Figure 8. Depiction of biotinylated molecules sequestered in a hydrogel format. Biotinylated cargo is released from scaffolds by dissociating it from the streptavidin protein and release profiles are tuned.

[0029] DETAILED DESCRIPTION

[0030] A hydrogel platform was developed that out-competes the current clinical standard (Dynabeads™) for human T cell expansion by over 70%. This scaffold is the first hydrogel platform to reach expansion throughputs on the order of 2000x, greater than the standard for ex vivo expansion, namely, Dynabeads™. This hydrogel is useful to enhance cell proliferation or to sequester and release biotinylated drugs on demand.

[0031] In certain embodiments, the hydrogel platform is composed of primarily PEGDA and PEGDA-streptavidin polymers. These PEGDA-strep hydrogels have tunable mechanical properties and can be used to present biotinylated ligands with tunable stiffness to cells. In certain embodiment, this system is used to expand human T cells for mock adoptive immunotherapy applications. In certain embodiments, the system is used to expand specific subsets of cells by selectively tuning immobilized ligands and other cues (such as cytokine mileu).

[0032] In one aspect, the expansion of CD4+ and CD8+ T cells has been performed (Example 1 below).

[0033] In certain aspects, provided herein is a hydrogel with orthogonal control of mechanics and ligand presentation comprising (a) polyethylene glycol diacrylate (PEGDA), polyethylene monoacryate (PEGMA), and / or Poly(N-isopropylacrylamide) (PNIPAM), (b) streptavidin acrylamide, and (c) a photoinitiator.

[0034] In certain aspects, the PEGDA has a molecular weight (MW) of 500 to 50,000 (or any value therebetween). In certain embodiments, the PEGDA has a MW of 500, 600, 700, 800, 900, 1000, 2000, 3000, 3400, 4000, 5000, 6000, 7000, 8000, 9000, 10,000, 15,000, 20,000, 30,000, 35,000, 40,000, 45,000 or 50,000.

[0035] In certain aspects the PEGDA is MW 500 to MW 35,000.

[0036] In certain aspects, the PEGDA is MW500, MW700, MW3400 or MW35000.

[0037] In certain aspects, the PEGDA is MW3400.

[0038] In certain aspects, the PEGDA present at a concentration of 1-100% (or any value therebetween). In certain embodiments, the PEGDA is present at a concentration of 1, 2, 3,4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, 30, 35, 40, 45, 50, 55, 60, 65, 70, 75, 80, 85, 90, 95 or 100%. In certain aspects, the PEGDA is present at a concentration of 4-25%. In certain aspects, the PEGDA is present at a concentration of 10-20%.

[0039] In certain aspects, the PEGMA is present at a concentration of 0-30%. In certain aspects the PEGMA is absent. In certain aspects, the PEGMA is present at a concentration of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30%. In certain aspects, the PEGMA is present at a concentration of 25%.

[0040] In certain aspects, the PNIPAM is present at a concentration of 0-30%. In certain aspects, the PNIPAM is present at a concentration of 1, 2, 3, 4, 5, 6, 7, 8, 9, 10, 11, 12, 13, 14, 15, 16, 17, 18, 19, 20, 21, 22, 23, 24, 25, 26, 27, 28, 29, or 30%.

[0041] In certain aspects, the photoinitiator is Lithium phenyl-2,4,6-trimethylbenzoylphosphinate (LAP), Irgacure 2959, Ruthenium or tetramethylethylenediamine (TEMED) in combination with sodium persulfate.

[0042] In certain aspects, the streptavidin acrylamide is present at a concentration of 0.001-5 mg / mL. In certain aspects, the streptavidin acrylamide is present at a concentration of 0.001, 0.002. 0.003, 0.004, 0.005, 0.006, 0.007, 0.01, 0.02, 0.03, 0.04, 0.05, 0.06, 0.07. 0.08, 0.09, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7. 0.8, 0.9, 1, 2, 3, 4, 5, mg / mL.

[0043] In certain aspects, the streptavidin acrylamide is present at a concentration of 100-200 pg / mL.

[0044] In certain aspects, the streptavidin acrylamide is present at a concentration of 2 mg / mL.

[0045] In certain aspects, the LAP is present at a concentration of 0.025-2.5%. In certain aspects, the LAP is present at a concentration of 0.025, 0.05, 0.075, 0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7. 0.8, 0.9, 1.0, 1.1, 1.2., 1.3, 1.4, 1.5, 1.6, 1.7, 1.8, 1.9, 2.0, 2.1, 2.2, 2.3, 2.4, or 2.5%.

[0046] In certain aspects, the LAP is present at a concentration of 0.05-0.2%.

[0047] In certain aspects, the LAP is present at a concentration of 0.1%.

[0048] In certain aspects, the hydrogel comprises molecular biology grade water.In certain aspects, hydrogel further comprises an additional co-polymer.

[0049] In certain aspects, the additional co-polymer is selected from the group consisting of Acrylic acid, acrylamide, bis-acrylamide, PEGMA and pNIPAM.

[0050] In certain aspects, the hydrogel further comprises a biotinylated molecule, wherein the biotinylated molecule operably links to the streptavidin.

[0051] In certain aspects, the hydrogel further comprises non-biotinylated bioactive molecules in the hydrogel (i.e., where the bioactive molecules are not cross-linked to the hydrogel).

[0052] In certain aspects, the biotinylated molecule or non-biotinylated bioactive molecule is a protein, small molecule, peptide or nucleic acid.

[0053] In certain aspects, the protein is an antibody.

[0054] In certain aspects, the antibody is specific for TCR, CD3, CD28, CD2, LFA-1, LFA3, PD1, PDL1, ICAM-1, ICOS, ICOS-L, IL2, IL17, chemokine receptor ligand, or IgG Isotope Control.

[0055] In certain aspects, hydrogel comprises (a) anti-human CD3, (b) anti-human CD28, and (c) anti-LFA-1, anti-CD2, or IgG Isotype Control are added in a ratio of 1 : 1 :2.

[0056] In certain aspects, the protein is TGF-B.

[0057] In certain aspects, the small molecule is a drug.

[0058] In certain aspects, the drug is blebbistatin, rapamycin, butyrate, or retinoid acid. In certain aspects, the nucleic acid is an aptamer.

[0059] In certain aspects, the composition comprises a solid substrate coated with the hydrogel described above.

[0060] In certain aspects, the solid substrate is a microsphere, a wire, a sheet, a mesh, or a supported or coated 3D structure.

[0061] In certain aspects, the microsphere is a magnetic component.

[0062] In certain aspects, the magnetic component is iron oxide. In certain aspects the material contains additional magnetic components (i.e., magnetic micro / nanoparticles that are cross-linked within the hydrogel framework).

[0063] In certain aspects, provided herein is a method of making a functionalized hydrogelthin film comprising:

[0064] a. consecutively treating a glass slide with 0.1 M NaOH, 2% 3-aminopropyltrimethoxysilane (APTMS), and 0.5% glutaraldehyde, with washing / drying steps in between to form a silane-functionalized slide,

[0065] b. mixing PEGDA, a photoinitiator, and streptavidin acrylamide in molecular biology grade water to form a hydrogel precursor solution,

[0066] c. placing a 20-200 pL droplet of the hydrogel precursor solution between a first silane-functionalized slide and a second silane-functionalized slide to form a sandwich, d. exposing the sandwich to light,

[0067] e. incubating the sandwich in PBS at 4°C to 20°C for 1- 24 hours,

[0068] f. removing one of the silane-functionalized slides from the sandwich to expose a thin film gel,

[0069] g. sterilizing the thin film gel,

[0070] h. contacting biotinylated molecules to the thin film gel to form a hydrogel substrate,

[0071] i. incubating the hydrogel substrate were incubated for about 1-16 hours at about 4-37°C with shaking, and

[0072] j . washing the hydrogel substrate with PBS and cell culture media to form a functionalized hydrogel thin film.

[0073] In certain aspects, the cell culture media is RPMI supplemented with 10% FBS, 1% sodium pyruvate, 55 pM beta-mercaptoethanol, lOO U / mL penicillin-streptomycin.

[0074] In certain aspects, the glass slide is a microscope coverslip.

[0075] In certain aspects, the coverslip is 12 mm x 12 mm.

[0076] In certain aspects, the PEGDA is 3400 MW PEGDA monomer, and is present at a concentration of 25%.

[0077] In certain aspects, the photoinitiator is LAP monomer, and is present at a concentration of 25%. In certain aspects, the streptavidin acrylamide, and is present at a concentration of 2 mg / mL.

[0078] In certain aspects, the light is at a wavelength of 250-800 nm.

[0079] In certain aspects, the light is at a UV wavelength of 250-420 nm.

[0080] In certain aspects, the light is at a visible wavelength of 400-800 nm.

[0081] In certain aspects, the light is at a UV wavelength of 365 nm.In certain aspects, the sandwich is exposed to the light for 10 seconds to one hour. In certain aspects, the sandwich is exposed to the light for 5-20 minutes.

[0082] In certain aspects, the sandwich is exposed to the light for about 10 minutes.

[0083] In certain aspects, the sterilizing is accomplished by immersing the gel in 100% ethanol.

[0084] In certain aspects, the gel is immersed in alcohol for about 30 seconds to 5 minutes. In certain aspects the gel is immersed in alcohol for about one minute.

[0085] In certain aspects, the biotinylated antibodies are mouse anti-human CD3, CD28, and LFA-1, CD2, or IgG Isotype Control in a ratio of 1:1:2 (anti-CD3: anti-CD28: anti-adhesion receptor / control). In certain aspects, the biotinylated antibodies are in a ratio of 1 : 1 : 1 or 1:1:5.

[0086] In certain aspects, the hydrogel substrates are incubated for 1-16 hours at about 4-37°C with shaking.

[0087] In certain aspects, provided herein is a functionalized hydrogel thin film made by the process of described above.

[0088] In certain aspects, provided herein is a method of expanding T cells comprising, seeding cells onto the functionalized hydrogel thin film substrates of described herein.

[0089] In certain aspects, the cells are primary human T cells.

[0090] In certain aspects, the cells are seeded at 5E5 cells / mL in 500 pL on functionalized PEGDA-Strep substrates on slides in 24-well plates.

[0091] In certain aspects, the method further comprises counting the cells and splitting the cell population to 2.5E5 cells / mL every 2 days starting on Day 3 after seeding until Day 5.

[0092] In certain aspects, the method further comprises removing the substrates on Day 5 and culturing the cells in suspension until day 13.

[0093] In certain aspects, the method further comprises counting the cells every 2 days splitting the cell population to 2.5E5 cells / mL until Day 13.

[0094] In certain aspects, the method further comprises isolating the cells by centrifugation to form an expanded cell product.

[0095] Certain embodiments of the invention will now be illustrated by the following nonlimiting Examples.EXAMPLE 1

[0096] Tunable enhancement of T cell expansion through modulation of stiffness and adhesion receptor engagement in an engineered hydrogel platform Abstract

[0097] Adoptive T cell therapies (ACT) are an important class of oncology treatments that require ex vivo T cell expansion for clinical success. Technologies which can control both phenotype and yield in expanded cell products are highly desired. Here, we develop a new hydrogel scaffold for controlled T cell expansion with yields of up to 2000x fold in two weeks, compared to other hydrogel constructs (~250x) and Dynabead™ magnetic beads (~1200x). Our 2D polyethylene glycol diacrylate (PEGDA) hydrogel scaffold is cross-linked with streptavidin moieties to present various biotinylated ligands to cells with controlled hydrogel stiffness (PEGDA- Strep). Using this platform, we demonstrate that combining substrate stiffness with adhesion receptor ligands (aLFA-1 or aCD2) dictates T cell activation and proliferation. On stiff substrates, these ligands drove expansions 49% (aLFA-1) and 68% (aCD2) greater than Dynabead™ magnetic beads with comparable T cell products, preceded by elevated metabolic and transcriptional activity. Notably, while stiff substrates increased yield, soft substrates produced T cells with superior antigen-specific killing selectivity. These findings highlight the role of mechanical sensing in T cell-APC interactions and suggest improved manufacturing methods for adoptive T cell therapy (ACT).

[0098] 1. Introduction

[0099] Adoptive immunotherapies, including chimeric-antigen receptor T cells (CAR) and tumor-infiltrating lymphocyte (TIL) therapies, have revolutionized the treatment of hematological malignancies. Current FDA approved therapies require the activation and expansion of T cells ex vivo in a 2-3 week manufacturing process, before cells are reinfused into patients. Emerging evidence suggests that differences in these activation and expansion steps can significantly alter resulting cell function. Accordingly, numerous biomaterial approaches have been developed to study key parameters in T cell activation and expansion. However, platforms that can induce T cell expansion on clinically relevant scales (i.e. -500-fold expansion in 2 weeks) and surpass the expansion potential of the clinical standard (Dynabead™ magnetic beads) have not yet been achieved.

[0100] Most of the biomaterials designed for ex vivo T cell activation to date have focused on presenting anti-CD3 and anti-CD28 ligands as stimulatory signals for T cell activation. These act as ligands for CD3 (Signal 1) and CD28 (Signal 2) on T cells; engagement activatesdownstream signaling associated with the T cell receptor (TCR). In the clinic, a polymer microparticle presenting signal 1 and 2 (Dynabead™ magnetic beads) has been used extensively for T cell expansion, whereas clinical T cell expansion on hydrogel substrates has been limited. Previous work has demonstrated T cell activation via anti-CD3 / CD28 is impaired on softer hydrogel substrates (ranging from l-20kPa) compared to stiffer substrates (ranging from 25-40kPa). Others have shown that T cell morphological and migratory characteristics, as well as cytokine production, are significantly altered in response to increases in stiffness. Several works have reported enhanced activation and expansion of T cells using hydrogel systems with optimized mechanical properties. For example, one report found that T cell fold expansion can be increased by nearly 3x using a hydrogel scaffold with an optimal stiffness of 25 kPa, compared to softer or stiffer conditions. Additionally, optimization of factors like oscillatory movement and ligand spacing can lead to enhancement of T cell activation, and this is mediated through molecular-scale force interactions of T cell surface proteins. For example, incorporation of mechanical oscillation during T cell activation has been shown to increase T cell proliferation by 1.7x. To date, the hydrogel system capable of the highest human T cell fold expansion is an alginate-based microgel platform that enables 500-700x fold expansion in 14 days. Other systems include a separate alginate microgel platform that enables 40-fold human T cell expansion in 6 days and a hyaluronic acid-based hydrogel scaffold that allows for 25-fold mouse T cell expansion in 7 days.

[0101] While most of these platforms rely solely on Signal 1 and Signal 2 engagement for T cell activation and expansion, several other ligand-receptor interactions also play a role in the induction of cell-wide programming responses. Two of these interactions are a) LFA-1 with ICAM-1, and b) CD2 with LFA-3. LFA-1 is a member of the integrin family found on the T cell surface, and it binds ICAM-1 found on lymphocytes, monocytes, and endothelial cells. LFA-l / ICAM-1 interactions allow for the generation of Signal 1 at low antigen concentrations by promoting the initial APC-T cell contact and strengthening it upon activation. LFA-1 interactions are also associated with extending duration of T cell-APC contact, inflammatory cytokine secretion, and T cell motility. CD2 is a costimulatory T cell surface glycoprotein that binds to LFA-3 expressed on antigen-presenting cells and is associated with the actin cytoskeleton. The presentation of LFA-3 has been reported to decrease the antigen threshold for T cell activation by 100-fold. Stimulation of CD2 through binding with LFA-3 can lead to outcomes similar to those of T cell activation, namelyphosphotyrosine signaling, calcineurin activity, IL-2 production and cell proliferation.

[0102] Despite early work showing increased T cell proliferation in the presence of LFA-1 or CD2-targeted costimulation, no biomaterial system has rigorously incorporated these signals or investigated them in the context of the mechanical microenvironment for T cell expansion.

[0103] Many hydrogel platforms for studying the effects of stiffness and ligand presentation on cell behavior have relied on polyacrylamide gels, which can cause cytotoxicity. Others have also developed hydrogel scaffolds using naturally derived materials, like alginate, collagen, and hyaluronic acid. While these are relatively biocompatible, they also present other ligands to cells (such as RGD) which may elicit cellular responses independently of mechanical properties and functionalized ligands. In this work, we develop a polyethylene glycol diacrylate (PEGDA) hydrogel scaffold with independent control of substrate mechanics and substrate ligand presentation. We incorporated streptavidin acrylamide within the gel which allows for immobilization of biotinylated ligands. Using this platform, we investigate the role that costimulatory adhesion receptor engagement plays during primary human T cell activation and expansion. We show that engagement of adhesion receptors LFA-1 and CD2, in combination with substrate stiffness, dictates early-stage activation and proliferation outcomes of primary human T cells. Additionally, longer-term T cell expansion can be enhanced by tuning both the substrate stiffness and adhesion ligand presentation profile of the activating scaffold. Surprisingly, although activation on stiffer substrates or with Dynabead™ magnetic beads led to increased cell yield compared to T cells activated on soft substrates, antigen-specific target cell killing was more selective in cells activated on soft substrates. We profiled the immunophenotypes of these expanded T cell products and analyzed how changes in early-stage transcriptional activity preceded increasing expansion due to stiffness and LFA-1 / CD2 engagement. This scaffold is, to our knowledge, the first hydrogel platform to reach expansion throughputs on the order of 2000x, greater than the clinical standard for ex vivo expansion, Dynabead™ magnetic beads.

[0104] 2. Results and discussion

[0105] 2.1 Development and validation of PEGDA-Strep scaffold with controlled mechanics and ligand presentation

[0106] We developed a hydrogel platform for the presentation of biotinylated ligands at various ligand densities with tunable substrate stiffness on a non-adhesive hydrogel. Briefly, poly-ethylene glycol diacrylate (PEGDA) was crosslinked with streptavidin acrylamide and polymerization was initiated with photoactivated LAP (Fig. 1A). The use of a cross-linkedstreptavidin protein within the hydrogel matrix facilitates inclusion of protein binding sites covalently incorporated into the hydrogel network (Figs. 1A, IB). Hydrogel mechanical properties can be altered by changing the concentration of PEGDA polymer. As such, this platform can be used to present any biotinylated ligand of choice on a non-adhesive PEGDA hydrogel, with controlled mechanical properties.

[0107] A silanized glass “sandwich gel” fabrication process was used to create a 2D thin film hydrogel, adapted from a protocol used for other hydrogel systems (Lee, K., et al., Matrix compliance regulates Raclb localization, NADPH oxidase assembly, and epithelial- mesenchymal transition. Molecular Biology of the Cell, 2012. 23(20): p. 4097-4108.). In brief, PEGDA-Strep hydrogel precursor was deposited between silanized and untreated glass coverslips and then cured under UV light. PEGDA monomer solutions between 5-20 w / v% were tested. Elastic modulus was measured by nanoindentation of the substrate and was relatively uniform throughout the gel. Elastic moduli values were 6 kPa, 60 kPa, 170 kPa, and 300 kPa for 5%, 10%, 15%, and 20% PEGDA concentrations, respectively (Fig. 1C).

[0108] Recently, the distinct effect of substrate elastic modulus vs substrate viscoelasticity on cell differentiation has been highlighted. For instance, several works have shown that T cell migration, expansion, and memory phenotypes can vary depending on the viscoelasticity of their environment. We performed constant strain stress relaxation tests on our PEGDA hydrogels. To complete a thorough characterization of the mechanical characteristics of PEGDA-Strep hydrogels, we next investigated the frequency- and time-dependent mechanical properties of the system. We found that there were equal, low amounts of stress relaxation across the stiffness range of interest for this study. A frequency sweep analysis confirmed that all PEGDA-Strep hydrogels tested exhibited loss factors lower than 0.15, indicating primarily elastic materials. Lastly, the creep strain was measured across all hydrogels, and the time-dependent creep strain for each gel followed similar profiles with strains of 40% or less after 2 minutes. This suggests our platform provides substrates with distinct elastic moduli and comparable time-dependent mechanical properties.

[0109] To functionalize PEGDA-Strep substrates with stimulatory antibodies for cellular assays, we added biotinylated antibodies on top of the gels using various combinations of anti-CD3, anti-CD28, anti-LFA-1, anti-CD2, or IgGl Isotype Control. Antibodies formed a functionalized layer on the surface of the hydrogel. We validated that stiffness of the PEGDA-Strep hydrogel did not alter the relative profile or amount of ligand binding / presentation (Fig. ID). There were no significant differences in the meanfluorescence intensity across stained gels of varied stiffness treated with the same concentration of ligand. The relative ligand density could also be reliably tuned by changing the ligand concentration log-fold (Fig. IE). We additionally verified that the relative densities of anti-CD3 (Signal 1) and anti-CD28 (Signal 2), were not significantly different between gels with various stiffnesses and ligand functionalization profiles of interest, and that and anti-IgG, anti-CD2 and anti-LFA-1 expression was not significantly different on gels of 6kPa to 60kPa (Figure S2A-E). This confirms the PEGDA-Strep hydrogel platform is a biorthogonal system for systematically investigating cellular responses to matrix stiffness and ligand presentation.

[0110] 2.2 Engagement of adhesion receptors on PEGDA-Strep hydrogels enhances early human T cell activation

[0111] We next performed T cell activation studies using anti-CD3 / CD28 in the presence or absence of additional adhesion ligands in a model human cell line (Jurkat T cell). Previous work has demonstrated T cell activation via anti-CD3 / CD28 is impaired on softer hydrogel substrates (ranging from l-20kPa) compared to stiffer substrates (ranging from 25-40kPa). Separately, it is also known that integrin receptor LFA-1 and CD2 co-stimulation can enhance T cell activation and expansion in suspension cell co-culture systems. However, whether the inclusion of these co-stimulatory ligands can “rescue” impaired activation commonly found on softer substrates is unknown. To test this, we incorporated LFA-1- and CD2-engaging signals into the PEGDA-Strep scaffold (Fig. 2A).

[0112] We first verified the biocompatability of the PEGDA-Strep platform with human CD4+T cells (Jurkat line) using a propidium iodide assay. Across all conditions over 95% of cells were alive with no significant difference shown from the untreated control. Jurkat cells were then cultured on hydrogel substrates coated with anti-CD3 / CD28 + / - LFA-1 or CD2 ligands. In the presence of anti-CD3 / CD28 / LFA-l, cells activated on stiffer substrates showed an increase in CD69 expression (from 22% to 63% positive) and a 6-fold increase in IL-2 secretion compared to cells treated on softer substrates with the same ligands. Similarly, in the presence of anti-CD3 / CD28 / CD2, CD69 expression increased (from 25% to 55% positive), accompanied by a 13-fold increase in IL-2 secretion on stiffer substrates.

[0113] Importantly, engagement of LFA-1 and CD2 via antibodies likely differs in both binding region and affinity compared to engagement via natural ligands including ICAM-1 and LFA-3. Accordingly, we verified that T cell activation was similar for both antibody-mediated and natural ligand-mediated engagement of these adhesion receptors in Jurkat cells. There wereno significant differences in CD69 expression or IL-2 secretion between cells treated with anti-CD3 / CD28 + ICAM-1 or anti-CD3 / CD28 + anti-LFA-1 at matched stiffnesses. This was the same for treatment with anti-CD3 / CD28 + LFA-3 and anti-CD3 / CD28 + anti-CD2. This suggests the stiffness effect seen here is less dependent on receptor binding affinity and validates the PEGDA-strep platform as a robust tool to study cell mechano-transduction with a variety of biotinylated ligands. Jurkat cells appeared mechanosensitive in the absence or presence of adhesion receptor engagement.

[0114] We next investigated how engagement of LFA-1 and CD2 alters primary human T cell activation. We explored the effects of both matrix stiffness and ligand density. Briefly, human CD3+T cells were isolated from healthy donors and cultured on PEGDA-strep hydrogels with varying stiffness (6, 60, or 170kPa) and ligand presentation (anti-CD3 / CD28, anti-CD3 / CD28 / LFA-l or anti-CD3 / CD28 / CD2). As expected, an increase in early activation markers (CD69, CD25 and IL-2 secretion) was observed in cells cultured on stiff 60 kPa substrates compared to softer 6kPa in all 3 metrics (Figs. 2B-2D). Importantly, engagement of adhesion receptors LFA-1 or CD2 generally increased the percentage of CD69+cells, CD25+cells and IL-2 secretion at 6kPa, 60kPa and 170kPa, compared to anti-CD3 / CD28 stimulation alone. On 60 kPa substrates, added LFA-1 engagement led to increased CD69 expression (from 40% to 59% positive) and CD25 (from 33% to 53% positive). IL-2 secretion was also increased by 2.5-fold across stiffness conditions compared to anti-CD3 / CD28 stimulation alone. Similarly, added CD2 engagement led to significant increases in CD69 expression (from 40% to 60% positive), CD25 expression (from 33% to 52% positive), and IL-2 secretion (2.3-fold increase) across stiffness conditions compared to anti-CD3 / CD28 stimulation alone. These findings suggest that simultaneous engagement of adhesion receptors promote a stronger activation phenotype and that T cell activation events (early activation marker expression and cytokine secretion) are still heavily dependent on substrate mechanics.

[0115] 2.3 Adhesion receptor engagement reduces stiffness-dependence of human T cell activation compared to classic stimulation alone

[0116] Next, we investigated if these effects were dependent on ligand density, stiffness, or a combination of these using orthogonal hydrogels. Briefly, hydrogels with controlled stiffness (6kPa, 60kPa and 170kPa), ligand density of O.lx, lx or lOx (validated by fluorescence microscopy, Fig 1H) and ligand presentation (anti-CD3 / CD28, anti-CD3 / CD28 / LFA-l or anti-CD3 / CD28 / CD2) were prepared (Fig. 3A). On lx anti-CD3 / CD28 substrates, cellscultured on substrates of 60 kPa and 170 kPa had significantly increased CD69 activation compared to the soft 6 kPa condition (Fig. 3B: from 8% to 39% positive for 60 kPa, from 8% to 44% positive for 170 kPa). This difference was still statistically significant at a 1 Ox ligand density. In fact, we found T cell activation was generally comparable on 60kPa and 170kPa substrates, with a notable decrease in CD69 expression observed in cells cultured on 6kPa substrates. While this relationship was also generally observed when anti-LFA-1 or anti-CD2 were added, the significance of the effect on soft substrates was reduced in the presence of adhesion ligands (Figs. 3C-3D). CD25 expression followed a similar pattern at both the lx and lOx ligand densities. On lx anti-CD3 / CD28 substrates, cells cultured on substrates of 60 kPa and 170 kPa had significantly increased CD25 activation compared to the soft 6 kPa condition (Fig. 3E: from 6% to 28% positive for 60 kPa, from 6% to 31% positive for 170 kPa). In conditions including adhesion receptor engagement, many of these increasing trends were no longer significant (Figs. 3F, 3G). This suggests that adhesion receptor engagement may reduce T cell sensitivity to substrate mechanics and better enable activation on softer substrates. Increasing substrate stiffness from 6kPa to 60kPa enhances T cell activation. Interestingly, previous work has reported that mouse lymph node stiffness increases in this specific range during the course of infection.

[0117] Similarly, T cells stimulated with anti-CD3 / CD28 alone (in the absence of adhesion receptor engagement) were sensitive to ligand density at higher stiffnesses (60 kPa and 170 kPa). In conditions including engagement of LFA-1 and CD2, the magnitude and significance of these ligand density-dependent increases are reduced. In some cases, the differences are no longer statistically significant, although a clear increasing trend exists. Taken together, our data suggests that T cells are mechanosensitive even in the presence of high concentrations of stimulatory and adhesion ligands. It has previously been established that engagement of adhesion receptors can alter the sensitivity of T cell activation to ligand density. Here, we show the first evidence that engagement of adhesion receptors can also reduce sensitivity to matrix stiffness, though this cannot completely “rescue” mechano-sensitivity.

[0118] 2.4 Adhesion receptor engagement and stiffness synergistically enhance T cell proliferation

[0119] Next, we investigated if this hydrogel platform could be used to enhance activation and expansion of primary human T cells, which is critical to the manufacturing of adoptive T cell therapies (ACT). Briefly, we isolated CD3+ T cells (mixed CD4+ and CD8+ cells) from healthy human donors and stimulated them on our hydrogel construct for 3-days. Cells werethen removed for analysis, or resuspended in tissue culture flasks for continued suspension culture for up to 14 days (Fig. 4A). On day 3, we imaged T cells in situ on hydrogels using brightfield and fluorescence microscopy. Cell adhesion (number of cells per area) was increased in the presence of added adhesion receptor engagement. Actin filament and microtubule formation was observed in the cells cultured on stiffer hydrogels, or in the presence of added adhesion receptor engagement, compared to softer 6kPa surfaces functionalized with only anti-CD3 / CD28 (Fig. 4C).

[0120] We assessed T cell proliferation by a carboxyfluorescein succinimidyl ester (CFSE) dye dilution assay and flow cytometry. At 3 days, proliferation was significantly increased on stiffer anti-CD3 / CD28 substrates (60 or 170 kPa) compared to the softer one (6 kPa), as has been previously reported. Importantly, the data showed that added engagement of LFA-1 or CD2 significantly increased proliferation compared to anti-CD3 / CD28 alone across all stiffness levels (Figs. 4D-4F). On 60 kPa gels, LFA-1 and CD2 engagement increased the proportion of proliferating cells to 84% and 80% (from 54%), respectively (Fig. 4E). CD25 expression was similarly enhanced by both stiffness and adhesion receptor engagement (Fig.

[0121] 4F). Comparing CD4+ and CD8+ T cells, there was no observed bias in initial proliferation. We next examined ex vivo expansion of primary human T cells activated on PEGDA-Strep substrates with varying stiffness and ligand presentation over clinically relevant timescales of two weeks (Figs. 4G-4I). For long-term primary human T cell expansions, cells were cultured on the hydrogel for 5 days, then transferred to a suspension culture for up to two weeks total. Importantly, cells in all conditions were adjusted to the same density on day 5 before continued culture in suspension, to avoid expansion bias from seeding density differences. At two weeks, T cell expansion was closely correlated with both initial activatory substrate stiffness and adhesion receptor engagement (Figs. 4G-4I). In general, treatment with softer 6 kPa PEGDA substrates led to decreased expansion compared to stiffer 60 kPa ones. Additionally, on substrates of both stiffnesses, inclusion of LFA-1 or CD2 engagement significantly increased fold expansion compared to the anti-CD3 / CD28 condition alone. On 6 kPa substrates, LFA-1 and CD2 engagement led to 5.6-fold and 6.1-fold increases in expansion, respectively, compared to anti-CD3 / CD28 alone. On 60 kPa substrates, LFA-1 and CD2 engagement led to 2.1-fold and 2.4-fold increases in expansion, respectively, compared to anti-CD3 / CD28 only on stiffer substrates.

[0122] The highest-expanding conditions were 60 kPa gels with adhesion receptor engagement, which significantly outperformed expansion with Dynabead™ magnetic beads,the current clinical standard (Fig. 41). Fold expansion yields are included in Table l.The average fold expansion following activation on 60 kPa PEGDA-Strep gels was 1840x with LFA-1 engagement and 2070x with CD2 engagement, compared to 1230x for Dynabead™ magnetic beads. In other words, T cell expansion on stiff gels with LFA-1 engagement led to a 49% improvement beyond Dynabead™ magnetic beads, and expansion on stiff gels with CD2 engagement led to a 68% improvement beyond Dynabead™ magnetic beads.

[0123] Interestingly, conditions in which both LFA-1 and CD2 ligands were included did not yield any further synergistic effect in either early activation or long-term expansion outcomes. To our knowledge, this is the first hydrogel system to achieve such enhancements in expansion (up to -2000-fold expansion) and surpass Dynabead™ magnetic beads controls.

[0124] Table 1. Summary of tunable fold expansion outcomes in response to varying stiffness and ligand presentation

[0125] Stiffness [kPa] Presented ligands Human T cell expansion (fold change)a)6 (PEGDA-Strep) aCD3 / aCD28 ~50

[0126] 6 (PEGDA-Strep) aCD3 / aCD28 / aLFA-1 -260

[0127] 6 (PEGDA-Strep) aCD3 / aCD28 / aCD2 -290

[0128] 60 (PEGDA-Strep) aCD3 / aCD28 -870

[0129] 60 (PEGDA-Strep) aCD3 / aCD28 / aLFA-1 -1840

[0130] 60 (PEGDA-Strep) aCD3 / aCD28 / aCD2 -2060

[0131] Dynabead™ magnetic aCD3 / aCD28c)-1230

[0132] beadsb)

[0133] a)Fold change from initial seeding density 5E5 cells / mL in 500 L, reported to 3 significant figures, mean of 6-8 donor expansions per condition;b,c)Commercially available Dynabead™ magnetic beads were used in this study as per manufacturers protocols (1 : 1 bead: cell ratio) as a positive control. These beads present aCD3 / aCD28; ligand ratio and concentration on Dynabead™ magnetic beads was not quantified.

[0134] 2.5 Evaluation of T cell functional activity using an antigen specific in vitro mouse model

[0135] Next, we investigated the functional activity of cells expanded using PEGDA-strep systems. As a model for tumor infiltrating lymphocyte adoptive cell transfer therapies, we developed an in vitro tumor cytotoxicity assay to assess TCR activated cell killing in cells expanded using our platform. To do this, we isolated CD8+ T cells from OT1 transgenic mice. OT-I T cells have a transgenic TCR that specifically recognizes the pMHC I presentingthe SIINFEKL peptide, a peptide moiety derived from ovalbumin. This peptide can be presented on H-2KbMHC Class I (MHC-I) expressing cells. As such, if H-2Kbmatched tumorigenic cell lines are pulsed with SIINFEKL, OT-1 T cells expressing the cognate TCR should be activated. OT-1 T cells can also be activated in a non-MHC restricted way using anti-CD3 / anti-CD28. We used a H-2Kbmatched Bl 6-F10-Luc target cell for this assay. Bib-Fl 0-Luc is an epithelial -like H-2KbMHC-I tumorigenic cell line with a constitutive luminescence reporter.

[0136] First, we adapted the hydrogel system to mouse cells by incubating PEGDA-strep gels of 6kPa and 60kPa with the relevant biotinylated ligands (anti-mouse-CD3, anti-CD28 and one of the following; isotype control antibody, anti-LFA-1 or anti-CD2). Next, we performed short term functional studies (Figure 5A-C). OT-1 CD8+ T cells were isolated from mice and cultured on PEGDA-Strep hydrogels of 6kPa or 60kPa for 24 hours in the presence and absence of stimulatory ligands. T cells were then removed from the gels, and co-cultured with B16-F10-Luc cells for an additional 24hrs. B16-F10-Luc cells were untreated or pulsed with SIINFEKL peptide for 6hrs prior to co-culture. OT-1 T cells cultured on unfunctionalized hydrogels did not kill non-SIINFEKL expressing target cells, as expected (Figure 5B).

[0137] Interestingly, target cell specific killing was enhanced -10% on cells cultured on hydrogel conditions compared to Dynabead™ magnetic beads controls (aCD3 / CD28: 24.6% Dynabead™ magnetic beads, 35.1% 6kPa, 34.6% 60kPa).

[0138] Next, we performed longer term expansion and cell killing assays with these cells (Figs. 5D-5G). OT-1 cells were expanded on the PEGDA-streptavidin system or in culture with mouse-specific Dynabead™ magnetic beads control particles for -10 days. As for human cells, mouse T cell expansion was significantly increased on 60kPa hydrogels presenting adhesion ligands (Fig. 5E). We found an ~8-fold improvement in T cell expansion compared to the Dynabead™ magnetic beads standard. Then, OT-1 CD8+ T cells were co-cultured with B16-F10-Luc target cells at a 10: 1 ratio and target cell killing was quantified. We found that across all conditions, all expanded T cells were highly effective at target cell killing (over 83%) and there were no significant differences in cytotoxic functionality across conditions (Fig. 5F). To investigate this further, we then performed a matched cell killing assay using OT-1 cells extracted from the expansion culture at day 1 and day 10, and co-cultured with B16-F10-Luc cells pulsed with SIINFEKL peptide at an intermediate effector to target cell ratio (5:1) (Fig. 5G). We found that on day 1, 60kPa hydrogel systems outperfomed both 6kPa and Dynabead™ magnetic beads with increasedtarget cell killing (aCD3 / CD28: 30.4% Dynabead™ magnetic beads, 34.3% 6kPa, 43.4% 60kPa). By Day 10, cell killing was significantly increased for all culture conditions, and differences between conditions were no longer statistically significant (aCD3 / CD28: 62.6% Dynabead™ magnetic beads, 54.6% 6kPa, 59.2% 60kPa).

[0139] We next tested the performance of expanded cells at different effector: tumor ratios, and analyzed off-target cell killing by comparing cell killing in pulsed and un-pulsed cells (Figs. 5H-5J). As expected, tumor cell killing was related to effectortumor ratio in all conditions, with higher effectortumor ratios leading to more killing in antigen-pulsed cells. Tumor cell killing could be observed at low effector to target ratios of 1 : 10 in vitro. As effector cell concentration was reduced, differences between cells expanded on 6kPa and 60kPa surfaces became amplified. For instance, cells expanded on anti-CD3 / CD28 surfaces co-cultured with pulsed target cell lines at an effectortarget ratio of 5: 1 showed target cell killing of 61.3% (Dynabead™ magnetic beads), 46.9% (6kPa hydrogel) and 57.0% 60kPa (hydrogel). Cell killing on softer substrates was further impaired at lower effector to tumor ratios of 1:10 (aCD3 / CD28: 39.4% Dynabead™ magnetic beads, 24.3% 6kPa, 35.1% 60kPa).

[0140] Interestingly, we also found significant differences in how cells respond to unpulsed cells linked to expansion protocol. Our data suggests cells expanded on softer substrates show lower killing overall, but enhanced antigen specificity compared to Dynabead™ magnetic beads or 60kPa substrates. Across all E:T ratios, cells expanded on 6kPa gels were slightly less efficacious at killing antigen pulsed cells (compared to Dynabead™ magnetic beads or cells expanded on 60kPa hydrogels with various ligands), however these cells showed much lower off-target killing of cells that were not pulsed with antigen (Figs. 5H-5J). This suggests in mouse cells, differences in T cell expansion protocol lead to differences in antigen specific vs general cytotoxicity, and that expansion on softer substrates may lead to optimal activation of antigen-specific T cells with fewer off-target effects. We next sought to characterize the phenotype of human T cells expanded on these substrates in depth.

[0141] 2.6 Immunophenotypic characterization of PEGDA-Strep T cell expansion products We characterized T cell subset and exhaustion phenotypes of expanded human cells (Fig. 6A). We first confirmed that viability of all expanded T cell cultures was higher than 90% using a propidium iodide exclusion assay. Comparing CD4 vs. CD8 T cell proportions of expanded T cells, in all conditions, the CD8+ T cell fraction was increased, as has been described by others. We found that T cell expansion using the clinical standard, Dynabead™magnetic beads, produced a mixed population containing -34% CD4+ and -66% CD8+ T cells on average (Fig. 6B). In general, expansion with our hydrogel system further shifted this towards a CD8+ dominant phenotype. For cells expanded following culture on softer substrates, the ratios of CD4+:CD8+cells were significantly lower (ranging from 0.20-0.27, Fig. 6B) than cells expanded on stiff substrates or with Dynabead™ magnetic beads (ranging from 0.35-0.43). This implies that initial activation on soft substrates and subsequent expansion leads to populations more heavily enriched in CD8+T cells. We believe this is the first reported work describing the impact of activatory substrate stiffness on CD4 vs. CD8 expansion bias.

[0142] We then investigated TH polarization phenotypes among the CD4+subset in T cell cultures expanded on the PEGDA-Strep system. It has been reported THI cells exhibit stronger cytotoxicity and cell-mediated immunity, while TH2 are more critical for humoral immunity responses. Previous works have established that cell surface markers can be used to distinguish these cell types (CD4+Tul-like cells; CXCR3, Tu2-like cells; CXCR3' CCR4+CCR6‘ , Tnl7-like; CXCR3 CCR4+CCR6+). Using this approach, we found conditions incorporating CD2 engagement exhibited significantly higher proportions of TH2 phenotypes (-30% on average) compared to below 20% in other conditions (Fig. 6C, S7D,E,F). Across all conditions, the proportion of cells exhibiting THI 7 was minimal (less than 3% on average).

[0143] A common measure of inflammatory T cell functionality is the ability to secrete multiple cytokines simultaneously upon activation. We explored the expression of various cytokines among the expanded T cell populations, namely IFNy, GzmB, IL-2, and TNFa. T cells expanded on soft gels without adhesion receptor engagement exhibited significantly higher IFNy secretion and lower IL-2 secretion than the other conditions (Fig. 6D). There were no significant differences in the proportion of cells that were capable of secreting 3 or more cytokines upon re-stimulation between conditions.

[0144] Next, we analyzed expression of T cell memory markers, first characterizing the proportion of T cell memory subsets (stem cell memory T cells (TSCm), central memory T cells (Tcm), effector memory T cells (Tem) and terminally differentiated effector memory cells re-expressing CD45RA (Temra)) (Fig. 6E). Previous works have established that TSCm-like phenotypes are associated with increased persistence and improved clinical outcomes, as well as enhanced survival and tumor control in preclinical models. T cells expanded on stiffsubstrates or soft substrates with adhesion ligands had significantly larger TSCm subsets (ranging from 19-22% of total) compared to soft anti-CD3 / CD28 substrates (13% of total, Fig. 6E). Though Dynabead™ magnetic beads-expanded T cells had a larger TSCm subset on average (17% of total) compared to the soft anti-CD3 / CD28 substrate, this was not statistically significant. Nearly all expanded T cells (>98%) expressed CD95 following activation and expansion, as has been previously established.

[0145] We also measured expression of CD127, a T cell stem memory marker on T cells. Similarly, CD127 expression was greater in T cell populations expanded on all stiff substrates or soft substrates with adhesion receptor engagement compared to the soft anti-CD3 / CD28 condition (Fig. 6F). Expression of CD27 is a costimulatory marker that is associated with T cell antitumor efficacy. We found that CD27 expression was also significantly increased in T cell populations expanded on stiff substrates or soft substrates with adhesion ligands (ranging from 66-70% CD27+) compared to the 6kPa anti-CD3 / CD28 substrate (21% CD27+) (Fig. 6G). Notably, T cells initially expanded on stiff substrates with anti-CD3 / CD28 (60kPa) or any substrate incorporating LFA-1 or CD2 engagement all have increased numbers of TSCm and CD27 expression compared to the current clinical standard, Dynabead™ magnetic beads. No differences were observed in expression of T cell exhaustion markers PD-1 and LAG-3 across conditions.

[0146] Taken together, this data suggests that careful selection of stiffness and presented adhesion ligands on hydrogel substrates during initial activation of T cells can alter subsequent expansion potential of specific cell phenotypes, including the CD4:CD8 ratio, ratio of memory / costimulatory phenotypes and ratio of THI and TH2 phenotypes. We next sought to validate these findings by investigating early timepoint transcriptional changes in human cells activated on hydrogel substrates.

[0147] 2.7 Adhesion receptor engagement and stiffness lead to distinct metabolic, mechanosensitive, and immune activation transcriptional program changes

[0148] Finally, we investigated gene expression changes which may explain differences in expansion. Human T cells were activated on soft / stiff PEGDA-Strep hydrogels or Dynabead™ magnetic beads controls and isolated after 3 days for bulk RNA-sequencing. Based on an initial clustering analysis, the samples were categorized into 3 main groups. These were A) cells activated using anti-CD3 / CD28 on soft substrates, B) cells activated using anti-CD3 / CD28 on stiff substrates, anti-CD3 / CD28 / LFA-l or anti-CD3 / CD28 / CD2 on soft substrates, and group C) cells activated using anti-CD3 / CD28 / LFA-l or anti-CD3 / CD28 / CD2 on stiff substrates or activated with Dynabead™ magnetic beads.

[0149] Additionally, as expected, untreated cells formed a separate fourth group in the complete clustering analysis. Interestingly, these groups corresponded with the general level of expansion observed earlier. Group A contained samples in the minimally expanding condition (soft gel without adhesion receptor engagement). Group B contained samples in moderately expanding conditions (soft gels with (or stiff gels without) adhesion receptor engagement). Group C contained samples in the maximally expanding conditions (stiff gels with adhesion receptor engagement or Dynabead™ magnetic beads controls). This pattern suggests that changes in transcriptional programming at an early timepoint of 3 days of activation correlated with expansion profiles.

[0150] Previous work has shown that variations in stiffness of the activating substrate leads to large numbers of differentially expressed genes (DEG). We performed DEG analysis to characterize significant variations in transcriptional programs linked to both stiffness and adhesion receptor engagement. We found 1646 genes were differentially expressed in T cells treated on soft vs stiff conditions when only anti-CD3 / CD28 ligands are presented.

[0151] Strikingly, comparing LFA-1 and CD2 engaging conditions to each other on either soft or stiff substrates, we found only 4-5 differentially expressed genes. In the soft condition, PTGIR, KLHL24, IFIT2, and IFI44 were all upregulated towards the LFA-1 engaging condition. For the most part, these genes have not been connected to T cell activation outcomes in the literature. However, signaling related to the prostacyclin receptor (PTGIR) has been identified as a positive regulator of both THI effector differentiation and CD8+T cell exhaustion. We observed higher levels of THI cells in cells expanded on substrates presenting LFA1 and increased relative levels of TH2 cells on CD2 substrates (Figs. 6C-6F), suggesting these early changes in transcriptional programming may provide insight into later phenotypic differentiation. This may also suggest that inclusion of LFA-1 or CD2 engaging ligands could be used to alter relative THI and TH2 polarization in other contexts. In the stiff condition, expression of TL22, TRANK 7, and TNIP3 were upregulated towards the LFA-1 engaging condition, while IL9 and MAO A were upregulated towards the CD2 engaging condition. IL-9 has been implicated in the proliferation and differentiation of T cells into Thl7 cells, however TH17 levels were low in all conditions. Cell yield also was slightly increased when using CD2 as a co-stimulatory ligand compared to LFA-1 in human T cell activation, but this was not statistically significant.We next performed a gene set enrichment analysis comparing the effects of added adhesion receptor engagement on transcriptional programs. We compared all samples to the soft 6 kPa anti-CD3 / CD28 control. Populations expanded on stiff substrates with LFA-1 or CD2 engagement were enriched for more gene sets overall (17 for LFA-1, 18 for CD2). Many of these are related to metabolism (e.g., cholesterol metabolism, amino acid degradation) and cell adhesion (e.g., cell adhesion molecules, lectin receptor signaling). A separate set of pairwise analyses also found that the number of enriched gene sets is preferentially increased in populations expanded on stiffer substrates or on substrates with adhesion receptor engagement. In general, we observe a pattern in which the number of enriched metabolism- and proliferation-related pathways based on differentially expressed genes increases as stiffness is increased and adhesion receptor engagement is added.

[0152] To further investigate the transcriptional differences between the varied PEGDA-Strep conditions, we analyzed relative gene expression of key gene sets, curated from the Gene Ontology gene set database, involved in T cell expansion, mechanosensing, and metabolism. Across all gene sets analyzed, there was very low expression in T cells expanded on soft conditions without adhesion receptor engagement. In populations expanded on soft substrates with LFA-1 / CD2 engagement or stiff substrates without these ligands, there was slightly higher expression of these genes. And in populations expanded on stiff substrates with adhesion receptor engagement, there was the highest expression of the genes of interest. A hierarchical clustering analysis based on the curated gene sets produced similar results to the clustering analysis performed after the principal component analysis. In short, these results show that increased T cell expansion is correlated with increases in transcriptional upregulation of genes to related T cell activation, mechanosensing, and metabolism.

[0153] 3. Conclusion

[0154] Here, we report a new biocompatible, biorthogonal hydrogel platform that can be easily functionalized with biotinytlated ligands to study cell behavior and expand various cell types. We used this platform for ex vivo T cell activation and expansion. Whereas most other biomaterial scaffolds are able to induce a fold expansion of up to ~250x, the PEGDA-Strep scaffold can induce two-week expansion levels up to 2000x under optimized substrate stiffness and ligand presentation conditions. Stiffer substrates (of at least 60 kPa or higher) yield optimal expansion levels compared to soft (6kPa). Ligand density can also be used to alter expansion profiles; overstimulation appears to lead to reduced expansion.We also use this system to explore the contribution of T cell adhesion receptor engagement to cell activation. We find added engagement of CD2 and LFA-1 both enhance activation marker expression and cytokine secretion during activation compared to the standard anti-CD3 / CD28 stimulation alone. Additionally, there is a lower dependence of activation on substrate stiffness when adhesion receptor engagement is present. This suggests for the first time that in addition to lowering density thresholds for T cell activation, adhesion receptors may also function to reduce mechanical stiffness thresholds for activation.

[0155] Crucially, we find that our optimal hydrogel system can be used to activate T cells before further expansion, leading to a T cell product that is functionally comparable to cells expanded by the current clinical standard, Dynabead™ magnetic beads, with around 50% increase in expansion yield. By controlling both the stiffness and ligand presentation of the activating PEGDA-Strep scaffold, we can tune the expansion outcomes between 50- and 2000-fold (Table 1). Incorporation of adhesion ligands into the activatory substrate further enhanced total expansion beyond Dynabead™ magnetic beads controls presenting Signal 1 and Signal 2 alone. Notably, while stiff substrates and Dynabead™ magnetic beads increased yield, soft substrates produced T cells with superior antigen-specific killing selectivity.

[0156] As the activatory substrate stiffness increases and adhesion receptor engagement is included, the expression of genes related to T cell activation, mechanosensing, and metabolism increases. GSEA also showed that pathways involved in T cell proliferation, immune pathway activation, and metabolism are enriched in conditions including adhesion receptor engagement on soft substrates, and further enhanced on stiff substrates. The present work further establishes the importance of tailoring both the biochemical and mechanical microenvironment of biomaterials during scaffold design to alter cell phenotype specific proliferation. We envision this platform and its future iterations to be useful in studying and improving T cell expansion through the modulation of mechanical and biochemical cues, or could be applied to study activation and expansion in a wide range of other cells.

[0157] 4. Experimental Methods

[0158] Hydrogel Synthesis and Characterization

[0159] PEGDA-Strep hydrogel synthesis:

[0160] PEGDA-Strep hydrogels were fabricated using a protocol adapted from Tse et al., “Preparation of Hydrogel Substrates with Tunable Mechanical Properties, Curr. Protcols in Cell Biology (2010). To alter stiffness of hydrogel scaffold, PEGDA monomer concentration was varied. Solutions of 25% 3400 MW PEGDA (Advanced Biomatrix, Cat. Cat. #GS705)and 2.5% LAP (Sigma Aldrich™, Cat. #900889) were made using molecular biology grade water. Hydrogel precursor solutions were made by mixing ratios of 25% PEGDA, 2.5% LAP, streptavidin acrylamide (Thermo Fisher™, Cat. #S21379), and molecular biology grade water as described in Supplementary Table 1. Hydrogels were tested for endotoxin and found to be endotoxin-free (<0.01 EU per gel). Gels were cured between 12 mm silane-functionalized and untreated coverslips under 365 nm UV light for 10 minutes. Silane-functionalized 12 mm glass slides were made by consecutive treatment with 0.1 M NaOH, 2% 3 -aminopropyltrimethoxy silane (APTMS), and 0.5% glutaraldehyde, with washing / drying steps in between. Gels were stored in PBS at 4 °C overnight prior to functionalization.

[0161] Prior to antibody functionalization, gels were opened and sterilized via immersion in 100% ethanol for 1 minute. Gels were then washed 3 times with endotoxin-free PBS. Then, the specified amounts of biotinylated mouse anti-human CD3 (Clone OKT3, Biolegend® Cat. #317320), CD28 (Clone CD28.2, Thermo Fisher™ Cat. #13-0289-82), andLFA-1 (Clone R7.1, Thermo Fisher™ Cat. #BMS102BT), CD2 (Clone RPA-2.10, Thermo Fisher™ Cat. #13-0029-82), or IgG Isotype Control (Clone P3.6.2.8.1, Cat. #13-4714-85) were added in a ratio of 1:1:2 (anti-CD3: anti-CD28: anti-adhesion receptor / control). Hydrogel substrates were incubated for 1.5 hours at room temperature with shaking, followed by washing with PBS and T cell media.

[0162] Mechanical characterization of PEGDA-Strep hydrogels:

[0163] The elastic modulus of the PEGDA-Strep hydrogels was measured using nanoindentation (Opticsl 1 Life® Pavone®). A 50- m spherical tip with nominal spring constant of 0.5 N / m was used to indent PEGDA-Strep hydrogels to a depth of 2 pm. Using the Hertzian contact model, the resulting load-indentation curve was analyzed to calculate the hydrogel’s elastic modulus (E). The following equation was used:

[0164]

[0165] where F is load, R is the tip radius, h is indentation depth, and v is Poisson ratio (assumed to be 0.5).Frequency- and time-dependent properties of the PEGDA-Strep hydrogels were measured using nanoindentation (Optics 11 Life® Pavone®). Stress relaxation analyses were conducted using nanoindentation with a 50-pm spherical tip with nominal spring constant of 0.5 N / m. Hydrogels were first indented to a load of 1pm and tip load was then recorded over 120 seconds at constant indentation. Four measurements were conducted per hydrogel sample and averaged together. Measured stresses were normalized to max stress and plotted as a function of time using RStudio. Frequency sweep analyses were conducted using nanoindentation with the same tip as above. Hydrogels were first indented to a depth of 1pm. Dynamic mechanical analysis was then conducted with an amplitude of 100 nm over 5 periods and a relaxation time of 2 seconds between frequency changes. Elastic modulus, loss modulus, and loss factor were calculated using DataViewer software (Optics 11 Life® Pavone®™). Creep compliance analyses were conducted using nanoindentation with the same tip as above. Hydrogels were first indented to a load of 0.5 pN and indentation depth was then recorded over 120 seconds at constant load. Four measurements were conducted per hydrogel sample and averaged together. Measured indentation depths over time were normalized to initial indentation depth and plotted as a function of time using RStudio.

[0166] Visualization of PEGDA-Strep hydrogel functionalization:

[0167] To determine relative densities of antibody attached to the substrates, substrates were functionalized with 4 pg / mL total mouse IgGl Isotype antibody, as was used for T cell activation studies. For O.lx and lOx densities, total antibody concentrations were 0.4 pg / mL and 40 pg / mL, respectively. The gels were then blocked with 3% BSA in PBS for 1.5 hours at RT with shaking. Next, they were stained with goat anti-mouse FITC secondary antibody in 3% BSA-PBS blocking solution for 1.5 hours at room temperature with shaking. Confocal microscopy (Zeiss™ Cell Discoverer 7) was used to measure fluorescence associated with hydrogel functionalization across conditions. To determine relative densities of specific anti-CD3 or anti-CD28 antibody attached to the substrates, substrates were functionalized with 1 pg / mL anti-CD3, 1 pg / mL anti-CD28, and 2 pg / mL anti-LFA-1, anti-CD2, or mouse IgGl Isotype Control for 1.5 hrs, as was used for T cell activation studies. The gels were then blocked with 3% BSA in PBS with 0.1% Tween 20 (PBS-T) for 1 hour at RT with shaking. Next, they were stained with 1 pg / mL of either His-tagged CD3 or His-tagged CD28 in 3% BSA-PBS blocking solution for 1 hour at room temperature with shaking. Lastly, the gelswere stained with 1 pg / mL anti-His tag FITC secondary antibody. In a separate experiment, anti-IgG, anti-CD2 and anti-LFAl were labelled with a protein labelling kit and tagged with AF488 dye. Proteins were purified by spin filtration. Confocal microscopy (Zeiss™ Cell Discoverer 7) was used to measure fluorescence associated with hydrogel functionalization across conditions.

[0168] Cellular assays

[0169] Cells and reagents:

[0170] Primary human T cells were isolated from peripheral blood leukapheresis product (Leukopaks sourced from StemCell™ Technologies). Jurkat T cells (RRID: CVCL 0065) were obtained at Passage 1 from Barker Hall Cell Culture Facility. T cells were cultured in T cell media: RPMI 1640 supplemented with GlutaMAX™ supplement, 10 mM HEPES (Gibco), 10% FBS (VWR™ Seradigm, Lot #156B19), 1 mM sodium pyruvate (Gibco), 55 pM B-mercaptoethanol (Gibco™), and 1% penicillin / streptomycin (Gibco™). B16-F10-Luc cells were cultured in DMEM supplemented with GlutaMAX™ supplement, 4.5 g / L D-Glucose, 10% FBS (VWR™ Seradigm), and 1% penicillin / streptomycin. Cells were passaged at 90% confluency and used at passage 20 or lower for all experiments. All cells were cultured under standard conditions (37 °C, 5% CO2).

[0171] Human T cell isolation and culture:

[0172] Jurkat T cells were frozen at Passage 2 in freezing media consisting of 50% FBS, 40% media, and 10% DMSO. Cells were thawed and allowed to propagate for 1 week prior to use in experiments. Cells were cultured in T cell media used until they reached Passage 20. Primary T cells were isolated using EasySep Pan-T Cell Isolation kit (StemCell™ Technologies) and subsequently frozen in freezing media consisting of 90% FBS and 10% DMSO. T cells were thawed and rested overnight prior to use in all experiments. Cells were mainly cultured in T cell media. For human T cell expansion experiments, T cell media supplemented with 30 lU / mL recombinant human IL-2 (Peprotech, Cat. #200-02). All cells were cultured under standard conditions (37 °C, 5% CO2).T cell early activation assay:

[0173] For early activation experiments, 5E5 cells / mL T cells were cultured for 20 hours on top of functionalized PEGDA-Strep substrates. The plate was placed on ice and cells were removed from wells with vigorous pipetting. Supernatants were collected and frozen at -80 °C for IL-2 ELISA quantification. Cells were initially stained with LiveDead™ Near-IR for 15 minutes at room temperature. Then, after washing, cells were stained with FITC-conjugated anti-CD69 (Clone FN50, Thermo Fisher™ Cat. #11-0699-42) antibody and PE-conjugated anti-CD25 (Clone BC96, Thermo Fisher™ Cat. #302606) for 1 hour on ice prior to flow cytometric analysis. IL-2 secretion was analyzed via enzyme-linked immunosorbent assay (ELISA) (IL-2 ELISA MAX kit, Biolegend®). For PEGDA-Strep biocompatibility testing, Jurkat T cells (50,000 in 0.5 mL) were seeded and cultured on unfunctionalized PEGDA-Strep gels in a 24-well plate. After 3 days of culture, T cells were stained with Hoechst 33342 (2 ug / mL, Thermo Fisher™ Cat. #H3570), propidium iodide (0.2 ug / mL, Sigma Aldrich Cat. #P4170) in FACS buffer (2% FBS and 1 mM EDTA in lx PBS). Data was acquired on a flow cytometer (BD Biosciences FAC Symphony™). For expanded primary human T cell viability testing, T cells were activated on PEGDA-Strep gels and expanded for 13 days prior to staining and flow acquisition as described above.

[0174] CFSE proliferation assay:

[0175] Cells were stained at 1E6 cells / mL with CellTrace™ CFSE dye (Thermo Fisher™, Cat. #C34554) diluted to 2.5 pM in PBS for 10 minutes at room temperature. 5 mL of media was added to cells, and cells were incubated for 5 minutes at 37 °C to quench unbound CFSE. Cells were washed twice and then plated on functionalized PEGDA-Strep substrates or other controls in a 24 well plate in 500 pL of cells at 5E5 cells / mL. After 3 days of incubation, cells were collected and proliferation peaks were analyzed via flow cytometry.

[0176] Confocal fluorescent and brightfield microscopy imaging:

[0177] For confocal microscopy fluorescent images, following 3 days of incubation on PEGDA-Strep substrates, T cells were stained with Hoechst 33342 (2 pg / mL, Thermo Fisher™ Cat. #H3570), CellMask™ Green Actin Tracking Stain (1:1000 dilution, Thermo Fisher™ Cat. #A57243), and Tubulin Tracker Deep Red (1:1000 dilution, Thermo Fisher™ Cat. #T34077) diluted in Hank’s Balanced Salt Solution (HBSS, Thermo Fisher™ Cat.#14025092). Cells were stained live for 30 minutes at 37 °C with lx probenecid to prevent reagent efflux, in accordance with manufacturer protocols, and gently washed with HBSS. Confocal fluorescent microscopy images were taken using the Zeiss™ CellDiscoverer 7 (inverted, 20x objective).

[0178] Brightfield microscopy images of unstained T cell cultures activated and unperturbed for 3 days on PEGDA-Strep substrates were also taken using the Zeiss™ CellDiscoverer 7 (lOx objective, 10 ms exposure, 10% light source intensity). Images were processed using ImageJ. Average number of cells per unit surface area was calculated by counting the number of cells in each of four 50 pm2square regions per donor per condition, averaging, and dividing by the area.

[0179] Long-term human T cell expansion and characterization:

[0180] Primary human T cells were seeded at 5E5 cells / mL on functionalized PEGDA-Strep substrates. For control condition, cells were mixed with Dynabead™ magnetic beads in a 1:1 ratio and used according to manufacturer recommendations as a positive control. Cells were counted and split to 2.5E5 cells / mL every 2 days starting on Day 3 after seeding until Day 13. Cells were removed from PEGDA-Strep substrates on Day 3-7 by vigorous pipetting. Cells were counted using Trypan blue exclusion on a Countess™ automated cell counter (Thermo Fisher™). The fold expansion for a given timepoint was iteratively calculated by dividing the timepoint’s counted density by the previously seeded density and multiplying by the fold expansion calculated for the previous timepoint.

[0181] Mouse T cell isolation and culture:

[0182] All procedures were conducted in accordance with National Institutes of Health (NIH) guidelines and were approved by UC Berkeley’s Institutional Animal Care and Use Committee (IACUC: AUP-2021-03-14192-1). Mice were purchased from Jackson Laboratory. 6-8 week old C57BL / 6-Tg(TcraTcrb)l lOOMjb / J (OT- 1) mice were used for mouse T cell expansion and functional cytotoxicity studies. Spleens were harvested and placed into 1 mL lx DPBS on ice. Spleens were mechanically dissociated and extruded through a 40 pm strainer to form a single cell suspension. The strainer was washed with 10 mL PBS. The suspension was treated with 5 mL lx RBC Lysis Buffer (eBioscience™) for 6 min before neutralization with 25 mL RPMI containing 10% FBS (VWR) and 1% Pen-Strep(Gibco™). The suspension was spun down, decanted, and resuspended to the desired cell concentration in cell culture media.

[0183] CD8+T cells were isolated from the splenocytes using a CD8a+MACS isolation kit (Miltenyi Biotec, Cat. #130-104-075). Mouse T cells were cultured in T cell media was supplemented with 20 U / mL (500 lU / mL) murine IL-2 (Peprotech®, Cat. # 212-12).

[0184] Long-term mouse T cell expansion:

[0185] Primary mouse T cells were seeded at 1E6 cells / mL in 500 pL on functionalized PEGDA-Strep substrates in 24-well plates. For control condition, cells were mixed with Dynabead™ magnetic beads in a 1:1 ratio and used according to manufacturer recommendations as a positive control. Cells were counted and split to 5E5 cells / mL every 2 days starting on Day 2 after seeding until Day 11. Cells were removed from PEGDA-Strep substrates on Day 2 by vigorous pipetting. Cells were counted using Trypan blue exclusion on a Countess automated cell counter (Thermo Fisher™). The fold expansion for a given timepoint was calculated as described above for the human T cell expansion experiments.

[0186] In vitro T cell cytotoxicity assay:

[0187] 1E4 target cells (B16-F10-Luc) were pulsed with 2 pg / mL SIINFEKL peptide in 100 uL media (DMEM + 10% FBS + 1% pen / strep) in 96-well flat-bottomed plates for 6 hours at 37 °C. Expanded effector OT-I T cells were then incubated with 1E4 pulsed or 1E4 non pulsed target cells at various ratios (1:10, 1:1, 5:1 or 10:1 E:T ratio) in 100 uL mouse T cell media for 24 hours. ONE-Glo EX reagent (100 uL, Promega™, E8130), was added to each well and the luciferase signal was measured after 5 minutes using a Tecan plate reader, in accordance with manufacturer protocols. The percent target cell killing was calculated using the following formula: (live control well - sample well) / (live control well - dead control well). Live control wells contained only 1E4 target cells and no T cells. Dead control wells contained 1E4 target cells that were treated with 50% ethanol 1 hour prior to assay readout.

[0188] T Cell Phenotypic Characterization:

[0189] For phenotyping characterization, on Day 13, cells were enumerated, stained using Live / Dead Near IR dye (Thermo Fisher™, L10119) for 10 minutes at room temperature, and fixed using 4% PFA on ice for 20 minutes. Cells were then stained with CD4-AF700 (CloneRPA-T4, Thermo Fisher™ Cat. #56-0049-042), CD8-PE-Cy7 (Clone RPA-T8, Thermo Fisher™ Cat. #50-112-9399), CCR7-BV605 (Clone G043H7, Biolegend® Cat. #353223), CD45RA-APC (Clone HI100, Biolegend® Cat. # 304112), CD127-BV421 (Clone A019D5, Biolegend® Cat. # 351310), CD27-FITC (Clone 0323, Biolegend® Cat. #302806), PD-1-PerCP-eFluor710 (Clone MIH4, Thermo Fisher™ Cat. #46-9969-42), and LAG-3-PE (Clone 3DS223H, Thermo Fisher™ at. #12-2239-42) antibodies diluted 1:100 in FACS buffer on ice for 30 minutes. For TH polarization studies, cells were stained with CD4-AF700, CD8-PE-Cy7, CXCR3-APC (Clone G025H7, Biolegend® Cat. #353707), CCR6-FITC (Clone G034E3, Biolegend® Cat. #353411), and CCR4-PE (Clone L291H4, Biolegend® Cat.

[0190] #359411) antibodies diluted 1 :200 in FACS buffer on ice for 30 minutes prior to flow analysis.

[0191] For cytokine secretion characterization cells were enumerated, stimulated with eBioscience™ Cell Stimulation Cocktail (Thermo Fisher™, 00-4970-93) in a 1:500 dilution and GolgiPlug protein transport inhibitor (BD Biosciences, 555029) at a 1:1000 dilution, and incubated for 5.5 hours. Following incubation, cells were stained using Live / Dead Near IR dye and fixed using 4% PF A. Cells were then stained with CD4-AF700, CD8-PE-Cy7, fFNy-APC (Clone 4S.B3, Thermo Fisher™ Cat. #50-153-22), GzmB-eFluor450 (Clone N4TL33, Thermo Fisher™ Cat. #48889642), TNFa-FITC (Clone MAbl 1, Thermo Fisher™ Cat. #50- 100-39), and IL-2-PE (Clone MQ1-17H12, Thermo Fisher™ Cat. #50-111-25) antibodies diluted 1 :300 in FACS buffer on ice for 30 minutes prior to flow analysis.

[0192] Flow cytometry:

[0193] Cells were stained with Live / Dead Near-IR (Thermo Fisher™, L10119) at a 1:1000 dilution in PBS for 10 minutes at room -temperature immediately prior to staining with fluorophore-conjugated antibodies for 30 minutes on ice. Flow cytometry data was collected on a BD Biosciences FACSymphony™ in the UC Berkeley Cancer Research Lab Flow Cytometry Facility. Single stain controls for compensation were prepared using UltraComp eBeads™ Plus Compensation Beads (Thermo Fisher™ Cat. #01-3333-42). A minimum of 10000 gated events was collected for each sample. Gating and data analysis was performed on FlowJo™ software.Bulk RNA Sequencing:

[0194] RNA was isolated from 3-day activated T cells using New England Biolabs® Monarch® Total RNA miniprep kit, according to the manufacturer’s instructions. Isolated RNA was analyzed for quality and quantity using Nanodrop. Isolated RNA was sent to Novogene Corporation Inc. (Sacramento, CA) for library construction, quality control, sequencing using the NovaSeq™ X Plus platform, and data filtering. Analysis of sequencing reads was conducted using NovoMagic cloud-based software. Principal component analysis clustering was performed using k-means clustering algorithm. Statistically significant upregulated DEG’s were identified using DESeq2 and chosen based on a statistical cutoff of |Log2FC| > 1 and p-value < 0.05. We first identified all differentially expressed genes in each of these comparisons that met a Log2FC threshold of 1 and padj threshold of 0.05. We then subjected significantly upregulated genes to a gene set enrichment analysis based on the Kyoto Encyclopedia of Genes and Genomes (KEGG) database. Next, we characterized the top 20 enriched KEGG gene sets across all conditions, and reported the significance values of each enriched gene set for each condition (Fig. 6e). For gene set enrichment analysis, upregulated DEG’s were analyzed using NovoMagic and the top 10 enriched gene sets for each condition with a padj < 0.15 were compiled into a list of gene sets to report in Fig. 6F.

[0195] Statistics:

[0196] All data are from at least three independent experiments and represented as mean ± SD unless otherwise stated. Significance testing was performed using one-way or two-way ANOVAs followed by post-hoc Student’s t tests for pairwise comparisons of interest.

[0197] Statistical testing was performed using GraphPad Prism software, and a significance level (a) of 0.05 was used for all tests. Figures were created using PRISM, BioRender and Adobe Illustrator.

[0198] EXAMPLE 2

[0199] Thin film substrate version of immunomodulatory hydrogel for T cell expansion PEG streptavidin acrylamide polymer blends were made, and UV cross-linked.

[0200] PEGDA streptavidin acrylamide or PEGMA streptavidin acrylamide with other co-polymers (i.e., pNIPAM) can also be made. In one embodiment, a standard PEGDA gel is a PEGDA monomer (25% 3400 MW PEGDA (Advanced Biomatrix, GS705)) and 2.5% LAP (Lithium phenyl-2,4,6-trimethylbenzoylphosphinate, Sigma Aldrich, 900889) and streptavidinacrylamide (2 mg / mL, Thermo Fisher, S21379) mixed in molecular biology grade water before UV curing. Next, biotinylated expansion ligands are added to the hydrogel. For example, mouse anti-human CD3, mouse anti-human CD28, and LFA-1, CD2, or IgG Isotype Control are added in a ratio of 1:1:2 (anti-CD3: anti-CD28: anti-adhesion receptor / control). In certain embodiments, the ratio is 1 : 1 : 1 or 1 : 1 :5. In certain embodiments, ligands of interest include ligands that bind to TCR, CD3, CD28, CD2, LFA3, PD1 / PDL1.

[0201] EXAMPLE 3

[0202] Thin film substrate version of immunomodulatory hydrogel for controlled drug delivery

[0203] As above, PEGDA monomer (25% 3400 MW PEGDA (Advanced Biomatrix, GS705)) and 2.5% LAP (Sigma Aldrich, 900889) and streptavidin acrylamide (2 mg / mL, Thermo Fisher™, S21379) were mixed in molecular biology grade water before UV curing. Then, biotinylated drug molecules were added to the hydrogel. The hydrogel can be used ex vivo or implanted in vivo as a drug depot. Drug molecules are released by administration of excess biotin. Kinetics of drug release can be tuned by controlled administration of biotin (or other streptavidin binding release molecules, i.e., ethanol).

[0204] EXAMPLE 4

[0205] Magnetic microbead substrate version of immunomodulatory hydrogel for ex vivo or in vivo manipulations

[0206] As above, PEGDA monomer (25% 3400 MW PEGDA (Advanced Biomatrix, GS705)) and 2.5% LAP (Sigma Aldrich, 900889) and streptavidin acrylamide (2 mg / mL, Thermo Fisher™, S21379) with a magnetic component (i.e., iron oxide) were mixed in molecular biology grade water or PBS. The solution was added to a surfactant stabilized oil phase and mechanically disrupted (i.e., sonication, vortex, microfluidic channel) to produce micro beads before UV curing. Then, biotinylated drug molecules were added to the hydrogel. The hydrogel can be used for cell expansion in culture, or implanted (and retrieved) for in vivo cell expansion or drug delivery. The inclusion of the magnetic particles renders the microbead hydrogel technology easy to remove from a mixture or from in vivo. Additional soluble drug molecules could be used. For instance, these beads could be implanted into the intraperitoneal space for the expansion of subsets of immune cells. Incertain embodiments, the hydrogel-coated microbeads are useful for in vivo expansion of T cells near internal organs in oncology applications.

[0207] EXAMPLE 5

[0208] Activation and expansion of subsets of human T cells for adoptive immunotherapy using a thin film immunomodulatory hydrogel with biotinylated expansion ligands anti- CD3, anti-CD28, anti-CD2

[0209] 1. Silane-functionalized 12 mm glass slides were made by consecutive treatment with 0.1 MNaOH, 2% 3 -aminopropyltrimethoxy silane (APTMS), and 0.5% glutaraldehyde, with washing / drying steps in between.

[0210] 2. PEGDA monomer (25% 3400 MW PEGDA (Advanced Biomatrix, GS705)) and photoinitiator (i.e., 2.5% LAP (Sigma Aldrich, 900889)) and streptavidin acrylamide (2 mg / mL, Thermo Fisher™, S21379) were mixed in molecular biology grade water.

[0211] 3. A 20-200 pl droplet of this solution was added to a 12 mm silanized-functionalized coverslip and sandwiched between two coverslips.

[0212] 4. Gels were exposed to 365 nm UV light for 10 minutes, then stored in PBS at 4°C overnight prior to functionalization.

[0213] 5. Gels were opened and sterilized. Then, biotinylated mouse anti-human CD3, CD28, and and LFA-1, CD2, or IgG Isotype Control were added in a ratio of 1 : 1 :2 (anti- CD3: anti-CD28: anti-adhesion receptor / control). Hydrogel substrates were incubated for 1.5 hours at room temperature with shaking, followed by washing with PBS and media.

[0214] 6. Primary human T cells are seeded at 5E5 cells / mL in 500 pL on functionalized PEGDA-Strep substrates on slides in 24-well plates. Cells were counted and split to 2.5E5 cells / mL every 2 days starting on Day 3 after seeding until Day 5.

[0215] 7. On day 5, substrates were removed, and cells were cultured in suspension until day 13. Cells were counted and split to 2.5E5 cells / mL every 2 days starting on Day 3 after seeding until Day 13.

[0216] 8. Final cell product is isolated by centrifugation and cell function is assessed before further downstream analysis and therapeutic application.EXAMPLE 6

[0217] In one aspect, FoxP3+ regulatory T cells have been enriched during expansion from primary blood mononuclear cells. In particular, tuning substrate properties (i.e., ligand presentation, stiffness) were used to enrich FoxP3 T cells from <10% to >40% of the population. Figure 7.

[0218] EXAMPLE 7

[0219] In certain embodiments, these hydrogels systems are useful for the controlled delivery of molecules. For instance, biotinylated molecules (proteins, small molecules, peptides, nucleic acid sequences, etc.) are sequestered in a hydrogel format. This hydrogel is freeze-dried for storage. Biotinylated cargo are released from scaffolds by dissociating it from the streptavidin protein (for instance, by flooding the scaffold with additional biotin). Release profiles could be tuned depending on the level of dissociation desired. In certain embodiments, the hydrogels are formed into various scaffolds formats (for example, coating wires or other materials, as a sheet, as a microsphere). In certain embodiments, a microsphere made of or coated with the hydrogel is implanted into the body and used to controllably deliver a biotinylated drug. Figure 8.

[0220] EXAMPLE 8

[0221] In certain embodiments, these hydrogels systems are coated onto a 3D architecture to increase surface area for cell expansion. For instance, a 2D or 3D lattice structure is produced from a filament polymer. This is coated with hydrogel precursor before cross-linking, then coated with biotinylated molecules of interest.

[0222] EXAMPLE 9

[0223] In certain embodiments, cells are encapsulated within the hydrogels and used for the production of bioactive molecules (i.e., proteins- B cells encapsulated within ligand coated hydrogels for continuous antibody production).

[0224] EXAMPLE 10

[0225] In certain embodiments, soluble biomolecules are embedded within these hydrogels to provide additional cues for expansion and differentiation (i.e., IL2, IL17, TGFb). In certain embodiments, systems are coated onto a 3D architecture to increase surface area for cellexpansions.

[0226] Although the foregoing specification and examples fully disclose and enable the present invention, they are not intended to limit the scope of the invention, which is defined by the claims appended hereto.

[0227] All publications, patents and patent applications are incorporated herein by reference. While in the foregoing specification this invention has been described in relation to certain embodiments thereof, and many details have been set forth for purposes of illustration, it will be apparent to those skilled in the art that the invention is susceptible to additional embodiments and that certain of the details described herein may be varied considerably without departing from the basic principles of the invention.

[0228] The use of the terms “a” and “an” and “the” and similar referents in the context of describing the invention are to be construed to cover both the singular and the plural, unless otherwise indicated herein or clearly contradicted by context. The terms “comprising,” “having,” “including,” and “containing” are to be construed as open-ended terms (i.e., meaning “including, but not limited to”) unless otherwise noted. Recitation of ranges of values herein are merely intended to serve as a shorthand method of referring individually to each separate value falling within the range, unless otherwise indicated herein, and each separate value is incorporated into the specification as if it were individually recited herein. All methods described herein can be performed in any suitable order unless otherwise indicated herein or otherwise clearly contradicted by context. The use of any and all examples, or exemplary language (e.g., “such as”) provided herein, is intended merely to better illuminate the invention and does not pose a limitation on the scope of the invention unless otherwise claimed. No language in the specification should be construed as indicating any non-claimed element as essential to the practice of the invention.

[0229] Embodiments of this invention are described herein, including the best mode known to the inventors for carrying out the invention. Variations of those embodiments may become apparent to those of ordinary skill in the art upon reading the foregoing description. The inventors expect skilled artisans to employ such variations as appropriate, and the inventors intend for the invention to be practiced otherwise than as specifically described herein.

[0230] Accordingly, this invention includes all modifications and equivalents of the subject matter recited in the claims appended hereto as permitted by applicable law. Moreover, any combination of the above-described elements in all possible variations thereof isencompassed by the invention unless otherwise indicated herein or otherwise clearly contradicted by context.

Claims

WHAT IS CLAIMED IS:

1. A hydrogel comprising (a) polyethylene glycol diacrylate (PEGDA), polyethylene monoacryate (PEGMA), and / or Poly(N-isopropylacrylamide) (PNIPAM), (b) streptavidin acrylamide, and (c) a photoinitiator.

2. The hydrogel of claim 1, wherein the PEGDA has a molecular weight (MW) of 500 to 50,000.

3. The hydrogel of claim 2, wherein the PEGDA is MW 500 to MW 35,000.

4. The hydrogel of claim 2, wherein the PEGDA is MW500, MW700, MW3400 or MW35000.

5. The hydrogel of claim 2, wherein the PEGDA is MW3400.

6. The hydrogel of any one of claims 1-5, wherein the PEGDA is present at a concentration of 1-100%.

7. The hydrogel of any one of claims 1-5, wherein the PEGDA is present at a concentration of 4-25%.

8. The hydrogel of any one of claims 1-5, wherein the PEGDA is present at a concentration of 10-20%.

9. The hydrogel of any one of claims 1-5, wherein the PEGDA is present at a concentration of 25%.

10. The hydrogel of any one of claims 1-9, wherein the PEGMA is present at a concentration of 0-30%.

11. The hydrogel of any one of claims 1-9, wherein the PNIPAM is present at a concentration of 0-30%.

12. The hydrogel of any one of claims 1-11, wherein the photoinitiator is Lithium phenyl- 2,4,6-trimethylbenzoylphosphinate (LAP), Irgacure 2959, Ruthenium or tetramethylethylenediamine (TEMED) in combination with sodium persulfate.

13. The hydrogel of any one of claims 1-12, wherein the streptavidin acrylamide is present at a concentration of 0.001-5 mg / mL.

14. The hydrogel of any one of claims 1-12, wherein the streptavidin acrylamide is present at a concentration of 100-200 pg / mL.

15. The hydrogel of any one of claims 1-12, wherein the streptavidin acrylamide is present at a concentration of 2 mg / mL.

16. The hydrogel of any one of claims 1-15, wherein the LAP is present at a concentration of 0.025-2.5%.

17. The hydrogel of any one of claims 1-15, wherein the LAP is present at a concentration of 0.05-0.2%.

18. The hydrogel of any one of claims 1-15, wherein the LAP is present at a concentration of 0.1%.

19. The hydrogel of any one of claims 1-18, wherein the hydrogel comprises molecular biology grade water.

20. The hydrogel of any one of claims 1-19, further comprising an additional co-polymer.

21. The hydrogel of claim 20, wherein the additional co-polymer is selected from the group consisting of Acrylic acid, acrylamide, bis-acrylamide, PEGMA and pNIP M.

22. The hydrogel of any one of claims 1-21, further comprising a biotinylated molecule, wherein the biotinylated molecule operably links to the streptavidin.

23. The hydrogel of any one of claims 1-22, further comprising a non-biotinylated bioactive molecule.

24. The hydrogel of claim 22 or 23, wherein the biotinylated molecule or non-biotinylated bioactive molecule is a protein, small molecule, peptide or nucleic acid.

25. The hydrogel of claim 24, wherein the protein is an antibody.

26. The hydrogel of claim 25, wherein the antibody is specific for TCR, CD3, CD28, CD2, LFA-1, LFA3, PD1, PDL1, ICAM-1, ICOS, ICOS-L, IL2, IL17, chemokine receptor ligand, or IgG Isotope Control.

27. The hydrogel of claim 26, wherein (a) anti-human CD3, (b) anti-human CD28, and (c) anti-LFA-1, anti-CD2, or IgG Isotype Control are added in a ratio of 1 : 1 :2.

28. The hydrogel of claim 24, wherein the protein is TGF-B.

29. The hydrogel of claim 24, wherein the small molecule is a drug.

30. The hydrogel of claim 29, wherein the drug is blebbistatin, rapamycin, butyrate, or retinoid acid.

31. The hydrogel of claim 24, wherein the nucleic acid is an aptamer.

32. A composition comprising a solid substrate coated with the hydrogel of any one of claims 1-31.

33. The composition of claim 30, wherein the solid substrate is a microsphere, a wire, a sheet, a mesh, or a supported or coated 3D structure.

34. The composition of claim 32, wherein the microsphere is a magnetic component.

35. The composition of claim 34, wherein the magnetic component is iron oxide.

36. A method of making a functionalized hydrogel thin film comprising:a. consecutively treating a glass slide with 0.1 M NaOH, 2% 3- aminopropyltrimethoxysilane (APTMS), and 0.5% glutaraldehyde, with washing / drying steps in between to form a silane-functionalized slide, b. mixing PEGDA, a photoinitiator, and streptavidin acrylamide in molecular biology grade water to form a hydrogel precursor solution,c. placing a 20-200 pl droplet of the hydrogel precursor solution between a first silane-functionalized slide and a second silane-functionalized slide to form a sandwich,d. exposing the sandwich to light,e. incubating the sandwich in PBS at 4°C to 20°C for 1- 24 hours,f. removing one of the silane-functionalized slides from the sandwich to expose a thin film gel,g. sterilizing the thin film gel,h. contacting biotinylated molecules to the thin film gel to form a hydrogel substrate,i. incubating the hydrogel substrate were incubated for about 1-16 hours at about 4-37°C with shaking, andj . washing the hydrogel substrate with PBS and cell culture media to form a functionalized hydrogel thin film.

37. The method of claim 36, wherein the cell culture media is RPMI supplemented with 10% FBS, 1% sodium pyruvate, 55 pM beta-mercaptoethanol, and 100 U / mL penicillin-streptomycin.

38. The method of claim 36 or 37, wherein the glass slide is a microscope coverslip.

39. The method of claim 38, wherein the coverslip is 12 mm x 12 mm.

40. The method of any one of claims 36-39, wherein the PEGDA is 3400 MW PEGDA monomer, and is present at a concentration of 25%.

41. The method of any one of claims 36-40, wherein the photoinitiator is LAP monomer,and is present at a concentration of 25%.

42. The method of any one of claims 36-41, wherein the streptavidin acrylamide, and is present at a concentration of 2 mg / mL.

43. The method of any one of claims 36-42, wherein the light is at a wavelength of 250- 800 nm.

44. The method of any one of claims 36-42, wherein the light is at a UV wavelength of 250-420 nm.

45. The method of any one of claims 36-42, wherein the light is at a UV wavelength of 365 nm.

46. The method of any one of claims 36-42, wherein the light is at a visible wavelength of 400-800 nm.

47. The method of any one of claims 36-42, wherein the sandwich is exposed to the light for 10 seconds to one hour.

48. The method of any one of claims 36-42, wherein the sandwich is exposed to the light for 5-20 minutes.

49. The method of any one of claims 36-42, wherein the sandwich is exposed to the light for about 10 minutes.

50. The method of any one of claims 36-49, wherein the sterilizing is by immersing the gel in 100% ethanol.

51. The method of claim 50, wherein the gel is immersed in alcohol is for about 30 seconds to 5 minutes.

52. The method of claim 50, wherein the gel is immersed in alcohol is for about oneminute.

53. The method of any one of claims 36-52, wherein the biotinylated antibodies are mouse anti-human CD3, CD28, and and LFA-1, CD2, or IgG Isotype Control in a ratio of 1:1:2 (anti-CD3: anti-CD28: anti-adhesion receptor / control).

54. The method of any one of claims 36-53, wherein hydrogel substrates are incubated for 1.5 hours at room temperature with shaking.

55. A functionalized hydrogel thin film made by the process of any one of claims 36-54.

56. A method of expanding T cells comprising, seeding cells onto the functionalized hydrogel thin film substrates of claim 55.

57. The method of claim 56, wherein the cells are primary human T cells.

58. The method of claim 57, wherein the cells are seeded at 5E5 cells / mL in 500 pL on functionalized PEGDA-Strep substrates on slides in 24-well plates.

59. The method of claim 57 or 58, further comprising counting the cells and splitting the cell population to 2.5E5 cells / mL every 2 days starting on Day 3 after seeding until Day 5.

60. The method of claim 59, further comprising removing the substrates on Day 5 and culturing the cells in suspension until day 13.

61. The method of claim 60, further comprising counting the cells every 2 days splitting the cell population to 2.5E5 cells / mL until Day 13.

62. The method of claim 61, further comprising isolating the cells by centrifugation to form an expanded cell product.

Images