Composite particle

Abstract

The invention relates to composite particles comprising a core of a hydrophobic phase, which may also be cross-linked, wherein the core is charged at its surface. A lipid bilayer is arranged on the surface of the core, enclosing it. This lipid bilayer comprises at least one lipid with a functional group. Correspondingly functionalized composite particles can be used to activate T cells.

Description

[0001] T-cell activation is a fundamental physiological process during inflammation, the antiviral response, and the formation of immune memory. It is also a crucial first step for the ex vivo expansion of T cells used for adoptive T-cell transfer in cancer immunotherapy (Waldman, AD, Fritz, JM & Lenardo, MJ. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nature Reviews Immunology 20, 651-668, (2020) doi:10.1038 / s41577-020-0306-5).

[0002] One of the current challenges limiting the wider application of such immunotherapeutic approaches is the reliable expansion of T cells with suitable phenotypes from a small, patient-derived initial population of immune cells. While this presents a significant challenge, it also offers an opportunity to explore the practical application of novel bioengineering concepts, such as synthetic bottom-up biology and synthetic cell engineering. The challenge lies particularly in the need to generate T cells in a phenotypic state compatible with effective therapies, e.g., non-exhausted cells with a lower self-antigen release threshold, increased resistance to immunosuppression, as well as memory and effector potency (Tantalo, DGM et al. Understanding T cell phenotype for the design of effective chimeric antigen receptor T cell therapies).Journal for ImmunoTherapy of Cancer 9, e002555, (2021) doi:10.1136 / jitc-2021-002555.). Fine-tuning the phenotypic composition of expanded T-cell populations is also a crucial prerequisite for tailoring T-cell products to the specific needs of different immunotherapeutic approaches (TCR-transduced T cells, chimeric antigen receptor T cells, or reinfused tumor-infiltrating lymphocytes (Zhang, DKY et al. Enhancing CAR-T cell functionality in a patient-specific manner. Nature Communications 14, 506, (2023) doi:10.1038 / s41467-023-36126-7)). With the use of antibody-decorated polystyrene beads (e.g., Dynabeads or MACSiBead) that crosslink and activate CD3, CD2, or CD28 receptors on T-cell membranes, simple ex vivo expansion of T cells has become a routine procedure (Li, Y. & Kurlander, RJ).Comparison of anti-CD3 and anti-CD28-coated beads with soluble anti-CD3 for expanding human T cells: Differing impact on CD8 T cell phenotype and responsiveness to restimulation. Journal of Translational Medicine 8, 104, (2010) doi:10.1186 / 1479-5876-8-104.). Such dispersed beads mimic in vivo signaling by antigen-presenting cells (APCs) and offer a controlled and reproducible technology for generating large numbers of T cells using biomimetic artificial APCs (aAPCs) (Isser, A., Livingston, NK & Schneck, JP Biomaterials to enhance antigen-specific T cell expansion for cancer immunotherapy. Biomaterials 268, 120584, (2021) doi:10.1016 / j.biomaterials.2020.120584.).The application of dispersed aAPC particles offers advantages over antibody-functionalized surfaces such as hydrogels or solid carriers, particularly due to their compatibility with 3D cultures and their easy scalability to therapeutic and industrial production systems.

[0003] Although cell-mimicking, solid, antibody-decorated spheres are capable of triggering crucial events in T-cell signaling, such as the activation of T-cell receptors and co-stimulation (signals I and II during T-cell activation) (Sckisel, GD et al. Out-of-Sequence Signal 3 Paralyzes Primary CD4(+) T-Cell-Dependent Immunity. Immunity 43, 240-250, (2015) doi:10.1016 / j.immuni.2015.06.023.), they do not provide important biophysical cues on APCs that control and fine-tune the T-cell expansion and differentiation process in vivo (Isser, A., Livingston, NK & Schneck, JP Biomaterials to enhance antigen-specific T cell expansion for cancer immunotherapy. Biomaterials 268, 120584, (2021) doi:10.1016 / j.biomaterials.2020.120584; Lambert, L.H. et al. Improving T Cell Expansion with a Soft Touch. Nano Lett 17, 821-826, (2017) doi:10.1021 / acs.nanolett.6b04071.).Biophysical features that aAPCs must mimic to achieve more programmable T-cell expansion include cell stiffness, ligand density, size, surface topography, and, most importantly, ligand lateral mobility. In this biophysical context, several studies highlight the importance of mechanosensitivity for T-cell stimulation and the role of aAPC stiffness in successful T-cell activation (Harrison, DL, Fang, Y. & Huang, J. T-Cell Mechanobiology: Force Sensation, Potentiation, and Translation. Frontiers in Physics 7, (2019) doi:10.3389 / fphy.2019.00045). Increased IL-2 production by CD4. +Mouse T cells were associated with increased stiffness of a stimulating polyacrylamide hydrogel (Judokusumo, E., Tabdanov, E., Kumari, S., Dustin, ML & Kam, LC Mechanosensing in T lymphocyte activation. Biophysical Journal 102, L5-7, (2012) doi:10.1016 / j.bpj.2011.12.011). On the other hand, substrates based on polydimethylsiloxane (PDMS) and human CD4 +A reverse trend has been observed in T cells (O'Connor, RS et al. Substrate rigidity regulates human T cell activation and proliferation. J Immunol 189, 1330-1339, (2012) doi:10.4049 / jimmunol.1102757). A more recent study suggests that these two seemingly contradictory trends may be linked and represent a biphasic mechanosensitive response (Yuan, DJ, Shi, L. & Kam, LC. Biphasic response of T cell activation to substrate stiffness. Biomaterials 273, 120797, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2021.120797). It is noteworthy that mechanosensitivity is also observed in other immune cells, such as... B. NK cells, was demonstrated where dispersed alginate particle substrates with different stiffnesses were used to assess mechanotransduction in natural killer cells (Friedman, D. et al. Natural killer cell immune synapse formation and cytotoxicity are controlled by tension of the target interface).Journal of cell science 134, (2021) doi:10.1242 / jcs.258570.). These studies demonstrate that stiffness plays a crucial role in modulating key aspects of T-cell activation and differentiation, highlighting the need for advanced α-APC technologies that effectively address a broader range of tissue stiffnesses, from 1 kPa (lymphatic tissue) to several hundred kPa (solid tumors) (Jansen, LE, Birch, NP, Schiffman, JD, Crosby, AJ & Peyton, SR Mechanics of intact bone marrow. J Mech Behav Biomed Mater 50, 299-307, (2015) doi:10.1016 / j.jmbbm.2015.06.023; Qin, Q., Wang, D., Xu, L., Lan, Y. & Tong, M. Evaluating Lymph Node Stiffness to Differentiate Bacterial Cervical Lymphadenitis and Lymph Node-First Presentation of Kawasaki). Disease by Shear Wave Elastography. Journal of Ultrasound in Medicine 40, 1371-1380, (2021) doi:https: / / doi.org / 10.1002 / jum.15518.).Perhaps even more important than the effects related to stiffness is the fact that previous aAPC approaches are based on immobilized and adsorbed ligands on solid, elastomeric or hydrogel supports, which are unable to mimic the dynamic, mobile lipid membrane environment and the membrane-membrane interface in the in vivo scenario. In particular, lateral mobility and spatial separation of receptors and ligands have been identified as critical factors for the formation of immune synapse (IS) architectures at the APC-T cell interface (Grakoui, A. et al. The Immunological Synapse: A Molecular Machine Controlling T Cell Activation. Science 285, 221-227, (1999) doi:10.1126 / science.285.5425.221; Davis, SJ & van der Merwe, PA The kinetic-segregation model: TCR triggering and beyond. Nature Immunology 7, 803-809, (2006) doi:10.1038 / ni1369).Immune synapses (ISs) are central regulatory structures that orchestrate the initial signaling events during T-cell expansion. Regulatory ligands and co-receptors such as PD-1, CD58, CTLA-4, CD28, or ICAM exhibit distinct spatial divisions into supramolecular clusters, regulatory contact zones, and receptor microclusters. These structures are essential for accurate antigen recognition, regulation of downstream signaling pathways, and prevention of unregulated interactions, thus ensuring controlled T-cell activation (Dustin, ML. The immunological synapse. Cancer Immunol Res 2, 1023-1033, (2014) doi:10.1158 / 2326-6066.Cir-14-0161). Importantly, the precise IS composition largely depends on lateral ligand mobility to enable the formation of spatially separated zones by antigen-producing cells (APCs) that control T-cell activation.To achieve a comparable level of biophysical fine-tuning in aAPCs, it is necessary to develop and implement this “membrane context” and the associated ligand mobility in particles, as these factors play a crucial role in mimicking the in vivo scenario and improving the efficiency and specificity of T-cell activation.

[0004] Recent advances in synthetic bottom-up biology have led to numerous new techniques for constructing artificial replicas of living cells, enabling the analysis of fundamental life processes under defined conditions (Hernandez Bücher, JE et al. Bottom-up assembly of target-specific cytotoxic synthetic cells. Biomaterials 285, 121522, (2022) doi:https: / / doi.org / 10.1016 / j.biomaterials.2022.121522; Jia, H. & Schwille, P. Bottom-up synthetic biology: reconstitution in space and time. Curr Opin Biotechnol 60, 179-187, (2019) doi:10.1016 / j.cop-bio.2019.05.008; Staufer, O. et al. Bottom-up assembly of biomedical relevant fully synthetic extracellular vesicles. Sci Adv 7, eabg6666, (2021) doi:10.1126 / sciadv.abg6666.). These synthetic cell technologies also have immense potential for advanced applications in biomedicine, as they closely mimic the structural and functional characteristics of living cells (Staufer, O. et al.Building a community to engineer synthetic cells and organelles from the bottom-up. eLife 10, e73556, (2021) doi:10.7554 / eLife.73556.). In principle, they could be used to build biomimetic systems that function as aAPCs and offer potential advantages due to their enhanced cell-like properties (Jenkins, E. et al. Reconstitution of immune cell interactions in freestanding membranes. Journal of cell science 132, (2018) doi:10.1242 / jcs.219709.). Since compartmentalization is a key feature of life, much of the work in synthetic bottom-up biology focuses on the development of well-controlled membrane systems (Göpfrich, K. et al. One-pot assembly of complex giant unilamellar vesicle-based synthetic cells. ACS Synthetic Biology, (2019) doi:10.1021 / acssynbio.9b00034.).This is a particular advantage when attempting to construct aAPCs, as their design requires a faithful replica of the cellular membrane interface (Céspedes, PF, Beckers, D., Dustin, ML & Sezgin, E. Model membrane systems to reconstitute immune cell signaling. Febs j, (2020) doi:10.1111 / febs.15488). One of the most widely used techniques for synthetic cell assembly is the use of oil-in-water or water-in-oil droplet templating (Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203). Such systems not only enable high-throughput production, but also offer finely adjustable production parameters and a wide range of molecular configurations.Since liquid droplet templating approaches are based on fully viscous oils, they also enable the production of ultra-soft dispersed substrates for T-cell expansion. In this context, previous studies have highlighted the use of elastomers such as PDMS for T-cell activation due to their low cost and high biocompatibility (O'Connor, RS et al. Substrate rigidity regulates human T cell activation and proliferation. J Immunol 189, 1330-1339, (2012) doi:10.4049 / jimmunol.1102757). Furthermore, it has been demonstrated that phospholipid bilayers can be deposited on cross-linked solid 2D PDMS substrates, enabling the formation of IS and the spatial separation of the ligands (Torres, AJ, Contento, RL, Gordo, S., Wucherpfennig, KW & Love, JC Functional single-cell analysis of T-cell activation by supported lipid bilayer-tethered ligands on arrays of nanowells. Lab on a chip 13, 90-99, (2013) doi:10.1039 / C2LC40869D; Lippert, AH et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal 120, 35-45, (2021) doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021.). The implementation of an oil elastomer-based bottom-up approach for synthetic cell assembly can therefore open up innovative avenues for the development of aAPC systems that better mimic the lipid membrane of natural APCs and integrate additional biophysical factors. Task

[0005] The object of the invention is to provide a system that can mimic the natural lipid membrane, as well as a method for producing such systems, and their use. Solution

[0006] This problem is solved by the inventions with the features of the independent claims. Advantageous embodiments of the inventions are characterized in the dependent claims. The wording of all claims is hereby incorporated by reference into the content of this description. The inventions also include all meaningful and, in particular, all mentioned combinations of independent and / or dependent claims.

[0007] The task is solved by a composite particle, comprising a) a core comprising a hydrophobic phase, wherein the core is charged at the surface; b) at least one lipid bilayer enclosing the core, and the lipid bilayer comprising at least one lipid with a functional group.

[0008] The core comprises a hydrophobic phase. This is understood to be a phase that is organic and immiscible with water; preferably, it is immiscible with water and comprises at least carbon atoms as well as hydrogen and / or fluorine atoms.

[0009] The hydrophobic phase is preferably an oil. An oil is defined as a hydrophobic substance that is liquid at a temperature of 25 °C and an atmospheric pressure of 1013 mbar. This applies in the uncrosslinked state. Oils based on hydrocarbons, which may also be partially or fully fluorinated, perfluorinated trialkylamines, or silicone oils are preferred. The oils may also be crosslinked.

[0010] A hydrophobic phase with a density between 0.9 g / mL and 1.2 g / mL is preferred.

[0011] A hydrophobic phase with a viscosity in the uncrosslinked state of 10 to 20000 cSt is preferred.

[0012] A silicone oil is understood to be an oil comprising at least one silicon atom, in particular at least one Si-O group, and most preferably an organopolysiloxane. Preferably, an organopolysiloxane comprising ethyl, phenyl, and / or methyl groups is used, preferably polyphenylmethylsiloxane or polydimethylsiloxane. Examples of silicone oils made from polyphenylmethylsiloxane are AR20 (Sigma-Aldrich, refractive index n20 / D 1.441–1.445, viscosity ~20 mPas (25 °C), density 1–1.02 g / mL at 20 °C) or AP200 (Sigma-Aldrich, refractive index n20 / D 1.502, viscosity ~100 mPas (25 °C), density 1.06–1.07 g / mL at 20 °C).

[0013] Examples of oils based on fluorinated or perfluorinated hydrocarbons are hydrofluoroethers such as CF3CH2OCF2H, CHF2CF2OCH3, CF3CHFCF2OCH3, CF3CH2OCH2CF3, n-C3F70CH3, C4F9OCH3, C4F9OC2H5, C7F 15 OC2H5.

[0014] Examples of trifluorinated trialkylamines are perfluorinated trialkylamines of the formula (C n F 2n+1)3N) n each has a value of 3 to 9, preferably 3 to 6, and the residues can be linear or branched. Examples are perfluorotri-n-butylamine, perfluorotri-n-pentylamine, or isobutylperfluoro-n-butylamine. Mixtures of the oils can also be used.

[0015] The hydrophobic phase, preferably the oil, and particularly preferably the silicone oil, can be cross-linked. This can be achieved by adding compounds modified with cross-linkable groups, for example, double bonds, vinyl groups, sulfide groups, epoxy groups, non-carbon-bonded hydrogen groups (e.g., silicon), alkoxy groups, and preferably vinyl groups. In the case of silicone oils, these can be polyorganosiloxanes modified with vinyl groups. These can react with each other and thus increase the stiffness of the hydrophobic phase. These compounds preferably have at least two cross-linkable groups per molecule. In the hydrophobic phase, preferably not all molecules are cross-linked, so that a specific viscosity is maintained.

[0016] The cross-linking process affects the viscosity of the composite particle.

[0017] The core of the composite particle is charged at its surface. The core can be positively or negatively charged. Preferably, the surface is negatively charged.

[0018] This is preferably achieved by incorporating at least one ionic surfactant into the core. During production, the charged groups align themselves on the surface of the core, thus forming a charged surface. The type of surfactant depends on the type of hydrophobic phase. A surfactant comprises at least one hydrophilic (polar or charged) part and at least one hydrophobic part. The surfactant is selected such that the hydrophobic part is soluble in the hydrophobic phase. In the case of silicone oils, the hydrophobic part preferably comprises alkyl and / or aryl groups. In the case of a hydrophobic phase consisting of fluorinated or perfluorinated hydrocarbons or perfluorotrialkylamines, surfactants with fluorinated or perfluorinated alkyl and / or aryl groups are preferred.

[0019] The amount of surfactant can also be used to control the size distribution of the nucleus.

[0020] The ionic surfactant is preferably an anionic surfactant. Particularly preferably, the anionic surfactant is selected from the group consisting of alkyl carboxylates, alkyl benzene sulfates, alkyl sulfonates, fatty alcohol sulfates, alkyl ether sulfates, sulfoacetates, or taurides; in the case of fluorinated surfactants, fluorinated or perfluorinated alkyl carboxylates or perfluoropolyether carboxylates such as poly(hexafluoropropylene oxide)carboxylate are used.

[0021] Examples of preferred anionic surfactants are lauryl sulfates, decyl sulfates, octyl sulfates, stearyl sulfates, and cetyl sulfates.

[0022] Preferably, alkali salts, in particular sodium salts of the anionic surfactants, are used, especially preferably sodium lauryl sulfate, sodium decyl sulfate, sodium octyl sulfate, sodium stearyl sulfate, sodium cetyl sulfate, and most preferably sodium lauryl sulfate.

[0023] To stabilize the core, it may be necessary to add a divalent cation, which also binds to the surface. A divalent cation is particularly preferably an alkaline earth ion, especially magnesium. This neutralizes the negative charge on the surface and stabilizes the particles.

[0024] In another embodiment of the invention, positively charged surfactants are used, which, for example, comprise quaternary ammonium groups.

[0025] The positively charged surfactants or cations result in a positive charge on the core's surface. This facilitates the bonding of the lipid bilayer to the core.

[0026] A lipid bilayer is arranged on the surface of the core. This means that the lipid bilayer surrounds the core and thus encloses it in a compartment. A unilamellar lipid bilayer is preferred.

[0027] The lipid bilayer preferentially comprises phospholipids.

[0028] Preferably, the lipid bilayer comprises phosphatidylcholine as a neutral amphiphile and at least one anionic lipid. The formation of such a lipid bilayer is also promoted by the presence of divalent cations.

[0029] Suitable phosphatidylcholines (PC) for the lipid bilayer include, but are not limited to, L-α-phosphatidylcholine, preferably from chicken egg (EggPC), 1,2-diacyl-sn-glycero-3-phosphocholine, 1,2-dimyristoyl-sn-glycero-3-phosphocholine (DMPC), 1,2-dihexanoyl-sn-glycero-3-phosphocholine (DHPC), 1,2-dipalmitoleoyl-sn-glycero-3-phosphocholine, 1,2-dipalmitoyl-sn-glycero-3-phosphocholine (DPPC), 1-palmitoyl-2-oleoyl-glycero-3-phosphocholine (POPC), and 1,2-distearoyl-sn-glycero-3-phosphocholine (DSPC). 1,2-Dilauroyl-sn-glycero-3-phos-phocholine (DLPC), and 1,2-dioleoyl-sn-glycero-3-phosphocholine (DOPC).

[0030] Suitable anionic lipids for the lipid bilayer include, but are not limited to, phosphatidylserines such as 1,2-dioleoyl-sn-glycero-3-phospho-1-serine (DOPS) and phosphatidylglycerols (PG) such as L-α-phosphatidylglycerol, preferably from chicken egg, 1,2-dioleoyl-sn-glycero-3-phospho-(1'-rac-glycerol) (DOPG), and phosphatidylinositols such as 1,2-dioleoyl-sn-glycero-3-phospho-(1'-myo-inositol) (DOPI).

[0031] Anionic lipids can be used as salts, especially as sodium salts.

[0032] The lipid bilayer may contain other components. Preferred additional components are phosphatidylethanolamines (PE).

[0033] For example, PEs may contain fluorescent dyes, such as Liss Rhod PE 1,2-dioleoyl-sn-gly-cero-3-phosphoethanolamine-N-(lissamine rhodamine B sulfonyl) (ammonium salt).

[0034] The lipid bilayer preferably comprises 60 to 80 mol% phosphatidylcholines and 10 to 35 mol% anionic lipids.

[0035] The lipid bilayer comprises at least one lipid, which includes at least one functional group. This can be directly linked to the lipid via a linker, for example, a linker comprising between 5 and 50 carbon atoms, where individual carbon atoms can be substituted with O, S, NH, C=O, -C(=O)O-, -C(=O)NH-.

[0036] The functional group is preferably a group through which coupling is possible (coupling group). These are groups that can selectively form covalent and / or coordinate bonds. Examples of such groups are epoxide groups, vinyl groups, hydroxyl groups, ester groups, carboxylic acid groups, thiol groups, carboxyl groups, biotin, streptavidin, nickel NTA, His tags, maleimide groups, amino groups, disulfides such as pyridyldithiopropionate (PDP), or N-hydroxysuccinimide groups (NHS). Particularly selective groups such as nickel NTA, biotin, maleimide groups, or N-hydroxysuccinimide groups are preferred.

[0037] An example of a nickel-NTA functionalized lipid is 1,2-Di-(9Z-octadecenoyl)-sn-glycero-3-[(N-(5-amino-1-car-boxypentyl)iminodieacetic acid)succinyl] nickel salt.

[0038] Examples of lipids with maleimide groups are 1,2-Dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-[4-(p-maleimidomethyl)cyclohexanecarboxamide] sodium salt, 1,2-Dioleoyl-sn-glycero-3-phos-phoethanolamine-N-[4-(p-maleimidomethyl)cyclohexanecarboxamide] sodium salt, 1,2-Dioleoyl-sn-glycero-3-phosphoethanolamine-N-[4-(p-maleimidophenyl)butyramide] sodium salt, 1,2-Dipalmitoyl-sn-glycero-3-phosphoethanolamine-N-[4-(p-maleimidophenyl)butyramide] sodium salt.

[0039] Examples of N-hydroxysuccinimide groups are palmitic acid N-hydroxysuccinimide esters and oleic acid N-hydroxysuccinimide esters. By using functionalized lipids, the density of modifications on the lipid layer can be controlled, which are selectively coupled to the functional group.

[0040] The viscosity of the particles results from the viscosity of the core, the oil-water interface, and the lipid bilayer. However, in the particles according to the invention, the viscosity of the core can be easily modified. This allows the overall elasticity of the particles to be controlled by changing the core.

[0041] The particles preferably have a diameter of less than 300 µm (determined optically on at least 10 particles), preferably from 1 to 200 µm.

[0042] The functional group of the particle, or the modification coupled to the functional group, can vary depending on the use of the particle.

[0043] The modification can also directly represent the functional group if correspondingly modified lipids are used.

[0044] In a preferred embodiment of the invention, the particles are modified with at least one antibody or antibody derivative. This antibody is preferably coupled to the lipid bilayer via the functional group, particularly preferably via a G-protein, and most preferably via a G-protein that is bound to a Ni-NTA lipid via a His-tag. In this case, the antibody or antibody derivative constitutes the modification.

[0045] A preferred use of the lipid particles according to the invention is the activation of T cells. In this preferred embodiment, the modification is a group for activating T cells, preferably a group with affinity for CD3, CD38, or CD28 receptors. The group can be an oligopeptide, protein, or antibody. Preferably, the group is at least one antibody, particularly preferably selected from the group comprising anti-CD3, anti-CD38, or anti-CD28 antibodies, particularly preferably human anti-CD3, anti-CD38, or anti-CD28 antibodies, or antibodies affine to modifications of these receptors.

[0046] Furthermore, modifications such as ligands or antibodies against the following receptors or surface proteins are preferred: T Cell Receptor (TCR), CD3 complex (CD3γ, CD3δ, CD3ε, and CD3ζ chains), CD28 (Co-stimulatory receptor), CTLA-4 (Cytotoxic T-Lymphocyte-Associated protein 4), PD-1 (Programmed cell death protein 1), LFA-1 (Lymphocyte function-associated antigen 1), ICOS (Inducible T-cell co-stimulator), CD2, CD4, CD8, CD5, CD6, CD7, CD27, CD40L (CD154), OX40 (CD134), OX40L, 4-1BB (CD137), 4-1BBL, CD30, CD30L, HVEM (Herpesvirus entry mediator), BTLA (B and T lymphocyte attenuator), PD-L1 (Programmed death-ligand 1), PD-L2, CD70, CD69, SLAM (Signaling lymphocytic activation molecule), CD44.

[0047] There can also be several modifications at the same time, preferably 1, 2, 3, 4 or 5 different ones, particularly preferably 1, 2 or 3, particularly preferably 1 or 2.

[0048] The density of functional groups, or modifications, on the surface of the composite particles preferably lies between 1 and 1000 functional groups / µm. 2, particularly preferably between 10 and 500 groups / µm 2 , especially preferred 20-300 groups / µm 2 .

[0049] Mixtures of several functional groups and / or modifications can also be used.

[0050] In a preferred embodiment, the elastic modulus of the composite particles lies between 0.5 kPa and 3 MPa, particularly preferably between 1 kPa and 2 MPa. With the particles according to the invention, it is particularly possible to produce particles with a stiffness in the lower range that is especially close to the stiffness of natural tissues.

[0051] The invention also relates to the use of the particles for the activation of T cells. For this purpose, the particles are modified accordingly, as described above. This use takes place particularly in vitro, for example by activating isolated T cells.

[0052] The invention also relates to composite particles which are modified with at least one tumor-associated antigen.

[0053] Examples of such antigens are Her2, ROR1, CD19, EGFRVIII, BCMA, CD20, CD33, CD123, CD22, CD30, CEA, EGFR, LMP1, LMP2A, mesothelin, PSMA, EpCAM, glypican-3, gpA33, GD2, and TROP2. These TAAs can be combined with the modifications described above to activate T cells. Their use is primarily in vitro.

[0054] The invention also relates to the use of the particles as cancer cell mimetics. For this purpose, the particles are preferably modified with at least one tumor-associated antigen as described above.

[0055] The invention also relates to the use of the particles according to the invention as a medicine.

[0056] The invention also relates to a method for producing the particles according to the invention.

[0057] This involves a procedure comprising the following steps: a) Providing a composition comprising at least one hydrophobic phase and at least one ionic surfactant; b) Producing an aqueous dispersion of the composition by generating oil droplets; c) optional addition of at least one divalent cation to stabilize the oil droplets; d) Addition of SUVs forming a lipid bilayer comprising at least one lipid with a functional group on the surface of the oil droplets.

[0058] The following section describes individual process steps in more detail. These steps do not necessarily have to be carried out in the order given, and the process described may also include further, unmentioned steps.

[0059] First, a composition comprising at least one hydrophobic phase and at least one ionic surfactant is provided. The embodiments described above are preferred.

[0060] The ionic surfactant is preferably an ionic surfactant as described above.

[0061] An aqueous dispersion of the composition is then produced by generating oil droplets. This is preferably done by mechanical action, particularly preferably via an ultrasonic bath. The dispersion is preferably an oil-in-water emulsion. The oil droplets correspond to the hydrophobic phase of the composite particles described above.

[0062] The surfactant preferentially adheres to the surface of the droplets.

[0063] Optionally, at least one divalent cation is then added to stabilize the oil droplets, especially if an anionic surfactant has been used.

[0064] In the next step, SUVs are added, forming a lipid bilayer comprising at least one lipid with a functional group on the surface of the oil droplets.

[0065] The functional group corresponds to the functional group as described for the composite particles.

[0066] The SUVs preferably comprise the lipid bilayer, according to the desired composition of the lipid bilayer of the particles to be produced. If the functional group is, for example, a Ni-NTA group, then this is included in the SUVs. The SUVs are preferably produced by extrusion. During mixing, the lipid bilayer adheres to the oil droplets.

[0067] It is also possible to add the divalent cations together with the SUVs. This can be done by adding two separate compositions.

[0068] The SUVs correspond with respect to the composition of the lipid bilayer as described above.

[0069] After the addition of the SUVs, the composition is preferably incubated for 2 to 60 minutes.

[0070] The particles are then preferably isolated by centrifugation. It may be necessary to resuspend the particles and centrifuge them again.

[0071] The particles can be stored as a suspension or dried. They are preferably stored as a suspension. Storage is preferably protected from light.

[0072] Preferably, the lipid bilayer comprises at least one lipid with a fluorescent group. This allows the lipid concentration to be determined. Concentration-dependent fluorescence measurements are performed for this purpose. This can also be done by comparing SUV solutions, in particular by comparing them with the same SUV solution that was used to deposit the lipid bilayer. The number of coupling sites can also be deduced from the fluorescence intensity.

[0073] The particles according to the invention can now be modified, in particular via their functional groups. In the case of antibodies, the Ni-NTA-modified particles are incubated with G-protein, to which the corresponding antibodies are then coupled. IgG antibodies are preferred, and IgG antibodies against human CD3, CD28, CD8, or modifications thereof, such as CD3-AlexaFluor488, are particularly preferred. The antibodies can also be added as a mixture with a defined ratio, or in the embodiments preferred above.

[0074] The modified particles are preferably purified by centrifugation to remove excess antibodies.

[0075] In a preferred embodiment, at least one crosslinking agent capable of crosslinking the hydrophobic phase is added in step a). Examples of such crosslinking agents are oils with crosslinking-capable groups, such as hydroxyl groups, vinyl groups, non-carbon-bonded hydrogen groups (e.g., silicon), alkoxy groups, or epoxy groups. The crosslinking-capable groups can be located at the ends or in the middle of the hydrophobic phase compounds. In the case of silicone oils, silicone oils with terminal groups are preferred. Vinyl groups are particularly preferred as crosslinking-capable groups. Adding a hardener may be necessary for crosslinking.

[0076] It may be necessary to further harden the droplets obtained in step b), for example by treating them at an elevated temperature, such as between 40 °C and 80 °C. This will harden the droplets.

[0077] Further details and features will become apparent from the following description of preferred embodiments in conjunction with the dependent claims. The respective features can be implemented individually or in combination with one another. The possibilities for solving the problem are not limited to the embodiments. For example, range specifications always include all intermediate values ​​not explicitly mentioned and all conceivable sub-intervals. Fig. 1. Structure of lipid bilayers on oil droplets. A) Schematic representation of the charge-mediated structure of bilayers on PDMS oil droplets. The droplets are stabilized by anionic surfactants containing divalent Mg. 2+-Cations bind to deposit an anionic lipid bilayer with fluorescent and NTA-functionalized lipids. (Oil droplet) B) and C) Representative scanning electron micrographs of PDMS droplets in phosphate-buffered saline (B) and in the dried state (C). Scale bars are 10 µm in B and 20 µm in C. (In solution; Dry) D) Representative confocal microscopy (Lipids) and brightfield images (BF) of the lipid layer that has formed around the oil droplets. Scale bars are 10 µm. E) Dynamic light scattering analysis of the dsLB zeta potential in the different assembly stages shown in A. The results are given as the mean ±SD of three repeated measurements. * p<0.05,**** p<0.001, one-sided ANOVA. F) Schematic representation of NBD fluorescence quenching by dithionite in lipid bilayers around the oil droplet.The membrane-impermeable dithionite only quenches the fluorophores in the outer membrane layer, resulting in a 50% decrease in intensity. G) Quantification of the fluorescence intensity of dsLBs with NBD-containing lipid membranes in a dithionite dilution series (Normalized number of NBD fluorophores). The results are presented as mean values ​​of two technical replicates. H) Representative confocal microscopy images of a FRAP experiment with dsLBs before and after bleaching, as well as after recovery. Scale bars are 5 µm (Atto 488 lipids). I) Fluorescence recovery quantification of FRAP series on dsLBs. The line shows the mean of n=9 dsLBs, the shaded line shows the standard deviation (Normalized fluorescent intensity = Normalized Fluorescence intensity; Time = Time);. Fig. 2. DsLB structure made of silicone oils. Representative confocal microscopy (rhodamine lipids) and bright-field images of the lipid layer that forms around droplets of silicone oils of different densities and viscosities. Scale bars are 30 µm; Fig. 3 DsLB assembly controls. Representative confocal microscopy (Rhodamine B) and bright-field images of dsLBs (top row) as positive controls and experiments with dsLBs with non-ionic surfactants (middle row) or without Mg. 2+ to form ions (bottom row). In the latter two cases, no circumferential lipid fluorescence is visible around the droplets, indicating unsuccessful dsLB formation. Scale bars are 100 µm; Fig. 4 Optimization of Mg 2+ and SDS concentrations for dsLB assembly. Representative confocal microscopy images of dsLBs assembled with different Mg concentrations. 2+and SDS concentrations were formed. The square image at 20 mM MgCl2 and 1 mM SDS shows the condition with the optimal combination of dsLB size, lipid layer intensity, and droplet quantity. Scale bars are 100 µm. Fig. 5. Optimization of SUV concentration for dsLB setup. A) Representative confocal microscopy (rhodamine B lipids) and bright-field images of dsLBs formed with different SUV concentrations (top). Scale bars are 100 µm. B) Analysis of the size distribution of the dsLBs shown in A (Count = number; Size = size). Fig. 6. Evaluation of the storage stability of dsLBs. Representative confocal microscopy (Rhodamine B) and bright-field images of dsLBs at three different magnifications after incubation in phosphate-buffered saline for 75 days at 4 °C. No changes in morphology, size, or lipid layer fluorescence were observed. Scale bars are 100 µm, 30 µm, and 10 µm for the left, center, and right columns, respectively. Fig. 7. CryoTEM analysis of dsLBs. A) Exemplary cryoTEM images of dsLBs. No glassy ice layer could be formed around the dsLBs, but membrane-like structures could be observed around the droplets (inset). The scale bars are 500 nm and 25 nm for the inset. B) Quantification of membrane thickness based on cryoTEM micrographs of 12 dsLBs and SUVs used for their fabrication. The results are given as mean ±SD; rectangular dots denote individual measurements on each dsLB membrane. Fig. 8. dsLB functionalization with recombinant proteins and antibodies. A) Schematic representation of dsLB functionalization with IgG antibodies via histidine-tagged (Hig-tag) protein G bound to NTA(Ni). 2+)-Lipids are coupled in the dsLB membrane. B) Representative confocal microscopy and bright-field images of the dsLB lipid layer (lipids) decorated with anti-CD3 AlexaFluor488 (antibody). The scale is 10 µm. C) Representative bright-field images of dsLBs formed from PDMS (top) and AR20 (bottom) silicone oils in RT-DC channels under deformation. D) RT-DC quantification of the dsLB elastic modulus (Young's modulus) for droplets formed from PDMS and AR20 oil. Results presented as median and 0.25–0.75 quartiles of >70 individual droplets. ****p < 0.0005, unpaired two-sided Student's t-test; Fig. 9. Calibration of dsLB antibody surface density. A) Representative bright-field and confocal microscopy with maximum Z projections of dsLBs (Rhodamine B) and MESF calibration particles (A488 MESF beads; dsLBs + Beads = dsLBs + Particles). The scale is 100 µm. B) Exemplary quantification of the fluorescence intensity of the MESF bead populations used for calibration (Blank; Count; Bead). C) Quantification of the IgG density on dsLBs produced with a 1:1 ratio of recombinant protein G to NTA lipids in the dsLB membrane (Count; AB Density). D) Calibration curve for deriving a function for the ratio of antibody density to protein G / NTA ratio; Fig. 10. Tuning of dsLB stiffness. A) Stereomicroscopic images of stiff (top row) and firm (bottom row) dsLBs, dried on plastic pipette tips at three magnifications. B) Representative confocal microscopy and brightfield images of the lipid layer forming around stiff and firm dsLBs. Scale bars are 10 µm. C) Quantification of the microscale Young's modulus of stiff and firm PDMS by nanoindentation analysis. Results are given as mean ±SD of 9 individual measurements. **** p<0.0005, unpaired two-sided Student's t-test. D) and E) FRAP analysis of the lipid layer mobility on stiff (D) and firm (E) dsLBs. The line shows the mean of n=10 (D) and n=9 (E) dsLBs, the shaded line shows the standard deviation (Normalized fluorescent intensity = normalized fluorescence intensity; Time = time); Fig. 11. Compression rheological analysis of PDMS material. Quantification of the elastic modulus (Young's modulus) of stiff and firm PDMS blocks (approx. 1 cm height and 1 cm diameter; bulk = bulk material) by compression rheology; Fig. 12. dsLB-mediated T-cell activation and expansion. A) Representative confocal microscopy (lipids) and brightfield (BF) images of T cells co-cultured with dsLBs. Scale bar: 10 µm. B) Confocal images of a contact event between a T cell and a rhodamine B-labeled dsLB decorated with fluorescent anti-CD3 antibodies. Scale bar: 5 µm. (Merge = superimposed) C) Brightfield microscopy image of an expanding T-cell cluster triggered by dsLB stimulation. Scale bar: 50 µm. D) Quantification of the expansion rate of T cells incubated with soft, firm, and stiff dsLBs, normalized to T cells expanded with DynaBeads. The results are given as the mean ±SD of two donors measured in duplicates.(Soft; firm; stiff; normalized expansion) E) CD25-expressing T-cell fractions of soft, firm, and stiff dsLBs compared to untreated, normal dsLBs and the DynaBeads control. Results are given as mean ±SD of four donors in duplicate. ** p<0.01 and ****p<0.001, one-way AnOVA assay (soft; firm; stiff; untreated; plain; fraction). Fig. 13. Phenotypic characterization of dsLB-expanded CD8+ T cells. A) PD1-expressing T-cell fractions from soft, firm, and stiff dsLBs compared to untreated and DynaBeads controls. Results are given as mean ±SD from four donors measured in duplicate samples. ** p<0.01 and ****p<0.001, one-way AnOVA assay. (soft; firm; stiff; untreated; fraction) B) Concentration of released cytokines from T cells expanded by soft, firm, and stiff dsLBs with 20 or 200 antibodies / pm 2on their membrane, each normalized to the DynaBeads-treated culture and compared to an untreated control. Results are presented as means of two donors in duplicate. (soft; firm; stiff; untreated; normalized concentration) C and D: Flow cytometric analysis of CD8+ T cells activated by soft dsLBs and stained for various surface markers. Untreated and DynaBeads-activated T cells are shown as controls. Data are presented as density plots for >2000 individual cells. Results were pooled from two donors and measured in duplicates (untreated). Fig. 14 The role of biochemical and biophysical factors in T-cell expansion. A) Representative bright-field and confocal microscopy images of silicon dioxide beads coated with a fluorescent lipid membrane. B) Flow cytometric quantification of CD25+ CD8+ T cells after expansion with different aAPC technologies. Results are given as mean ±SD of four donors in duplicate (soft; firm; stiff; untreated; plain). C) and D) Flow cytometric quantification of CD25+ CD8+ T cells (C) and expansion quantification (D) of CD8+ T cells after stimulation with aAPCs exhibiting different configurations for signals I, II, and III. Results are given as mean of four donors in duplicate. E) Schematic representation of the bispecific TcE stimulation setup with dsLBs.F) Representative confocal microscopy images of a primary CD8+ T cell incubated with bispecific TcE and dsLBs (lipids) presenting the TAA Her2 (merge). Scale bar is 10 µm. G) and H) Flow cytometry-based quantification of CD25+ (G) T cell populations, normalized to DynaBead-expanded and PD1+ (H) T cell populations in a TcE dilution series incubated with Her2-presenting dsLBs of varying stiffness. Results are given as mean ±SD of two donors in duplicate determinations. Fig. 15 A three-dimensional space for aAPC-based T-cell expansion. Schematic representation of the four aAPC technologies tested in this study via biochemical signaling strength, substrate stiffness, and lateral ligand mobility, illustrating the synergistic contribution of these three factors to T-cell expansion and the importance of lateral ligand mobility for a PD1-expressing phenotype (T cell activation; suppression susceptibility; low; high; moderate). Fig. 16 Integration of magnetic nanoparticles into dsLBs. A) Representative transmission electron micrograph of magnetic nanoparticles synthesized as described in the Materials and Methods section. B) Bright-field image of dispersed dsLBs without (left) and with hydrophobic magnetic NPs aligned in a magnetic field. C) Comparison of dsLBs with different concentrations of magnetic NPs before (left) and after (right) separation with a laboratory magnet. The percentages indicate the extraction efficiency of dsLBs as measured by light scattering. Fig. 17 Gating strategy for flow cytometric analysis of T-cell marker expression (untreated; viable cells; single cells); Fig. 18. Analysis of CD4 / CD8 T-cell populations after expansion. CD8 T cells were isolated by negative selection and either incubated untreated with plain dsLBs or expanded with soft dsLBs for 9 days. The cells were stained for CD4 and CD8 and analyzed by flow cytometry. The results were pooled from two donors and measured in duplicate (Untreated; Plain). Fig. 19. PD1 expression in memBeads and dsLB-expanded T cells. Flow cytometric analysis of PD1 expression in CD8+ T cells expanded for 9 days with dsLBs of varying stiffness or with memBeads. Results are presented as mean ±SD of two donors measured in duplicates (soft; firm; stiff). Fig. 20. Calibration of the Her2 surface density of dsLBs. Calibration curve derived from the MESF bead-based calibration of the Her2 density on dsLBs using the same procedure as in Fig. 9 shown (ratio = ratio).

[0078] Ex vivo T-cell expansion is crucial for effective immunotherapy but is currently limited by a lack of expansion approaches that accurately mimic in vivo T-cell activation. Drawing on synthetic bottom-up biology principles, we present a novel synthetic cell technology based on dispersed liquid-liquid phase-separated lipid bilayers (dsLBs) with tunable biochemical and biophysical properties as artificial antigen-presenting cells (aAPCs) for ex vivo T-cell expansion. Results obtained with the dsLB technology reveal three key findings: First, the introduction of laterally mobile stimulatory ligands onto soft aAPCs promotes the expansion of IL-4 / IL-10-secreting regulatory CD8+ T cells with a PD1-negative phenotype, which are less susceptible to immunosuppression.Second, it is shown that lateral ligand mobility can mask the differential T-cell activation observed on substrates with varying stiffness. Third, dsLBs were used to uncover a mechanosensitive component in bispecific Her2 / CD3 T-cell engager-mediated T-cell activation. Based on these three findings, it is proposed that lateral ligand mobility, alongside receptor and mechanosignaling, be considered a third crucial dimension for the development of ex vivo T-cell expansion technologies.

[0079] A synthetic cell technology based on droplet-supported lipid bilayers (dsLBs) has been developed, integrating mobile lipid membrane bilayers and tunable ligand densities in combination with adjustable and extremely low substrate stiffnesses as a novel aAPC technology. DsLBs are based on micrometer-sized silicone oil droplets generated by oil-in-water emulsions and coated with protein-functionalized phospholipid bilayer membranes. The use of anionic surfactants to stabilize the droplets enables charge-dependent deposition of lipid membranes onto the droplets, allowing for adjustable stiffness of the resulting lipid bilayers through varying degrees of crosslinking.The application of dsLBs for the activation of primary human T cells in vitro is demonstrated in comparison to commercially available DynaBead technology, and it is shown that the resulting T cell populations consist of a regulatory and memory-like phenotype. The addition of lateral membrane fluidity to the dsLB systems provides an additional layer of control, largely neglected by previous technologies, for fine-tuning the ex vivo expansion of T cells at therapeutic levels. Results

[0080] A synthetic cell construction strategy for dsLBs was developed, based on charge-mediated coating of PDMS oil droplets with lipid bilayers ( Fig. 1A). For this purpose, oil-in-water (O / W) droplets separated in the liquid-liquid phase are produced from PDMS / phosphate-buffered saline (PBS) emulsions and stabilized by anionic sodium dodecyl sulfate (SDS) surfactants. After the addition of divalent cations (Mg 2+ ) become small, negatively charged unilamellar vesicles (SUVs) that contain fluorescent (rhodamine B or Atto488) and NTA(Ni 2+ The droplet surfaces contain functionalized lipid head groups deposited to form unilamellar lipid bilayers. This manufacturing process, carried out under physiological buffer conditions, enables scalable mass production of biocompatible droplets with a spherical lipid membrane surface and a low-viscosity oil core.

[0081] In particular, it enables the production of soft dispersed aAPC substrates (~1 kPa) during T-cell activation and further crosslinking of the dsLB oil core to gradually increase its stiffness up to the MPa range, analogous to bulk PDMS substrates. 2.1. Formation and characterization of lipid bilayers using droplets

[0082] In a first step, a PDMS-based O / W emulsion (1:10 w / w ratio) was produced by sonication, which formed dsLBs according to the planned formation strategy. Investigation of the dsLB solutions by scanning electron microscopy (ESEM) revealed stable spherical droplets with smooth surface topography and similar contrast properties between the individual droplets ( Fig. 1B, C). Furthermore, even at extremely high, almost spherical packing densities, no fusion of the droplets was observed, proving their stability. Confocal microscopy of the assembled dsLBs revealed a fluorescent lipid layer surrounding the droplets, consistent with the inventive structure ( Fig. 1D). It is noteworthy that similar droplet structures could be composed with a variety of other silicone oils ( Fig. 2) To investigate the importance of charge-based interactions for dsLB assembly, the ionic surfactant SDS was replaced by the non-ionic surfactant Triton X-100 and it was found that no fluorescent lipid layer surrounds the droplet surface, indicating unsuccessful dsLB assembly ( Fig. 3 middle series). Similar results were obtained when attempts were made to synthesize dsLBs without Mg. 2+to form ions (bottom row of the figure), which led to droplets without a surrounding lipid layer. Furthermore, dynamic light scattering measurements of the zeta potentials of dsLBs were carried out at the different stages of assembly to evaluate a possible charge-mediated formation mechanism ( Fig. 1E). While “naked” PDMS droplets stabilized by anionic SDS exhibited a large negative zeta potential of -57 mV (±6 mV), the addition of MgCl₂ reduced the zeta potential to -45 mV (±1 mV). Subsequent incubation with SUVs further reduced the dsLB zeta potential to -26 mV (±1 mV), which is comparable to the value measured for the SUVs used for dsLB assembly (-19 mV (±3 mV)). Taken together, this suggests a sequential, layer-by-layer assembly of lipid membranes on oil droplet supports, driven by charge-mediated interactions between the individual building blocks.

[0083] To optimize the dsLB formation process and size distribution, the SDS and MgCl2 concentrations used for production were varied between 0.1 and 10 mM and 10 and 100 mM, respectively. Fig. 4) It was found that the optimal combination of dsLB size, lipid layer intensity, and droplet volume was achieved at 1 mM SDS and 20 mM MgCl₂. Furthermore, these conditions resulted in a mean dsLB size of 3.75 µm (± 2.61 µm), a range suitable for T-cell activation (Mescher, MF Surface contact requirements for activation of cytotoxic T lymphocytes. J Immunol 149, 2402-2405 (1992)), and a PDI of 0.58. Only SDS concentrations below 1 mM and MgCl₂ concentrations above 20 mM led to abnormal droplet morphologies with thick fluorescent lipid layers. The process of dsLB formation thus appears to be very stable over a wider range of SDS and MgCl2 concentrations, suggesting that dsLB formation is primarily driven by charge-based interactions that give the process robustness.It is noteworthy that the variation in SUV concentration added in the last step of the formation had no significant effect on the size distribution of the dsLBs or the production yield (. Fig. 5) In addition, the storage stability of the dsLBs in PBS at 4 °C was investigated over a period of 75 days to estimate the shelf life of the particles ( Fig. 6) Confocal microscopy analysis showed that the morphology, size, and lipid layer fluorescence of the dsLBs remained unchanged throughout the experimental period, indicating that the dsLBs are very stable in solution. Based on this optimization, 1 mM SDS and 20 mM MgCl₂ were defined as standard conditions for dsLB formation and applied in the experiments.

[0084] Confocal microscopy results showed the presence of a circumferential lipid layer on the oil droplet nucleus. However, it was not possible to derive information about the configuration of the lipid platelets from this, and in particular to rule out the formation of a monolayer with lipid fatty acid tails extending into the oil nucleus, as has been previously reported for other biomimetic droplet systems (Campàs, O. et al. Quantifying cell-generated mechanical forces within living embryonic tissues. Nat Methods 11, 183-189, doi:10.1038 / nmeth.2761 (2014).). To detect the presence of a lipid bilayer on the surface of dsLBs, previously developed NBD quenching assays were performed using lipids with NBD fluorophore-conjugated head groups (Ohmann, A. et al. A synthetic enzyme built from DNA flips 107 lipids per second in biological membranes. Nature communications 9, 2426, doi:10.1038 / s41467-018-04821-5 (2018).).This test uses the oxyanion dithionite, which effectively quenches NBD fluorescence but is impermeable to lipid membranes ( . Fig. 1F). Upon addition of dithionite to lipid bilayer membranes containing fluorescent NBD head groups, such as SUVs or presumably dsLBs, half of the NBD fluorescence is quenched, reflecting the uniform distribution of the NBD fluorophores among the lipids. Consequently, the fluorescence signal is completely lost in the presence of a lipid monolayer, whereas in multilayer membranes only a fraction of the original fluorescence is lost. Since this test is sensitive to the ratio of NBD fluorophores to dithionite molecules, the NBD fluorescence of dsLBs is quenched in a dithionite dilution series between 0.2 and 50 mM ( Fig. 1G). A plateau region was found between 6.25 and 50.0 mM dithionite, in which the dsLB fluorescence is attenuated to about 50% of the original fluorescence level. This quantitative analysis is a strong indication of the presence of a lipid bilayer on dsLBs. This finding is not only consistent with the charge-based formation model but is also confirmed by a cryogenic transmission electron microscopy (cryoTEM) analysis of the dsLBs ( Fig. 7) While attempts to form glassy ice layers around the dsLBs for cryoTEM analysis were unsuccessful, membrane-like structures on the dsLBs could be resolved, exhibiting a thickness similar to that of the SUVs used for their assembly. Furthermore, the analysis of fluorescence recovery after photobleaching (FRAP) revealed the lateral mobility of the lipids on the dsLB surface ( Fig. 1 H, I). Half-recovery rates of 12.2 seconds (±7.7 seconds) and mobile fractions of >0.72 (±0.15) were measured, which is in good agreement with the lipid diffusion characteristics in other supported lipid bilayer model systems, especially on PDMS supports (Lippert, AH et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal 120, 35-45, doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021 (2021)). Overall, this is compelling evidence that dsLBs are enveloped by a lipid bilayer membrane and are therefore suitable as a model for living cell membranes. 2.2 DsLB Functionalization and Stiffness Modulation

[0085] By installing NTA(Ni 2+ )-functionalized lipids were able to conjugate polyhistidine-labeled recombinant proteins, especially IgG antibodies, to the dsLB surface ( Fig. 8A,B). The quantitative conjugation of tunable antibody surface densities with fluorescent beads presenting known molecules of an equivalent soluble fluorochrome (MESF) was found ( Fig. 9A,B), calibrated. For this purpose, dsLBs were produced that were conjugated with recombinant his-labeled protein G to capture AlexaFluor488-conjugated antibodies on the dsLB surface ( Fig. 9C), and these samples were spiked with calibration beads. This is a modular and flexible approach that allows a variety of IgGs to be conjugated to the dsLB surface. Using confocal imaging and the fluorescence of the dsLBs and MESF beads, calibration curves were generated and regression parameters derived to determine the surface density of the antibodies on the dsLBs. With this approach, antibody densities in the range of 20–200 molecules / µm were achieved. 2 ( Fig. 9D) can be reliably varied, which is not only in the density range of natural ligands found on APCs (typically between 1 and 1000 ligands / µm). 2 ), but also in the densities used in other APC technologies (Schuster, H. et al. The immunopeptidomic landscape of ovarian carcinomas. Proceedings of the National Academy of Sciences 114, E9942-E9951, doi:10.1073 / pnas.1707658114 (2017); Hammink, R. et al. Semiflexible Immunobrushes Induce Enhanced T Cell Activation and Expansion. ACS Applied Materials & Interfaces 13, 16007-16018, doi:10.1021 / acsami.0c21994 (2021).). These results also show that dsLBs can present stimulating antibodies on their surface, which has proven to be a successful strategy for various other aAPCs, which typically use anti-CD3 and anti-CD28 antibodies to trigger initial activation and co-stimulation signals for T-cell expansion.

[0086] In a next step, the elastic properties of the dsLBs were characterized. Typically, aAPC technologies use particles and substrates with elastic moduli between ~10 kPa (e.g., hydrogels) and ~2 GPa (e.g., solid beads). Depending on the degree of crosslinking, the crosslinking time, and the temperature, the elastic modulus of PDMS elastomers can vary between ~100 kPa and 3 MPa (Hocheng, H., Chen, C.-M., Chou, Y.-C. & Lin, C.-H. Study of novel electrical routing and integrated packaging on bio-compatible flexible substrates. Microsystem Technologies 16, 423-430, doi:10.1007 / s00542-009-0930-2 (2010); Izdihar, K. et al. Structural, Mechanical, and Dielectric Properties of Polydimethylsiloxane and Silicone Elastomer for the Fabrication of Clinical-Grade Kidney Phantom. Applied Sciences 11, 1172 (2021).).Since the previously used dsLBs are based on an uncured, fully viscous core of PDMS droplets, their elastic modulus is expected to be close to 0 kPa. However, they still exhibit an elastic component due to the significant surface tension at the oil-water interface. Therefore, the elastic modulus of dsLBs is a product of the elastic properties of the oil core, which are mainly determined by the oil viscosity or degree of crosslinking, and the surface tension at the dsLB-water interface. The apparent elastic modulus of dsLBs was measured at the single-droplet level using high-throughput real-time deformation cytometry (RT-DC). Fig. 8C). For dsLBs based on PDMS oil with a viscosity of 5100 cP (hereinafter referred to as soft dsLBs), an average apparent modulus of elasticity of 3.85 kPa (± 1.60 kPa) was measured ( Fig. 8D). Interestingly, when producing dsLBs based on the silicone oil AR20, which has a lower viscosity of 20 cP, the effective elastic modulus of the dsLBs could be reduced to 1.34 kPa (± 0.32 kPa). Therefore, dsLBs with a non-crosslinked core exhibit elastic properties comparable to the lower range of physiological cell stiffness.

[0087] To increase the elastic modulus of the dsLBs core, the PDMS was crosslinked after emulsification but before the addition of MgCl2 and the formation of the bilayers. For this purpose, the PDMS droplets were cured in a ratio of 10:1 (stiff dsLBs) and 50:1 (rigid dsLBs) between PDMS and hardener (CA). Droplets were successfully produced for both stiff and rigid dsLBs, and these were even stable in air. Fig. 10A) and from which dsLBs with the characteristic circumferential lipid layer could be built ( Fig. 10B). The cured PDMS bulk material exhibited mean moduli of elasticity of 2990 kPa (± 8.94 kPa) and 5.75 kPa (± 0.66 kPa) for stiff and firm dsLBs respectively, as measured by compression rheology ( Fig. 11) However, the elastic properties of elastomers on the microscale can differ considerably from their bulk properties. Therefore, the microscale elastic properties were measured using nanoindentation analysis, and elastic moduli of 4.59 kPa (± 0.59 kPa), 11.08 kPa (± 2.95 kPa), and 163.46 kPa (± 64.33 kPa) were found for soft, firm, and stiff dsLBs, respectively. Fig. 10C). Remarkably, FRAP analysis showed that the dsLB membranes remained mobile on the hardened droplets, with mobile fractions of 0.90 (±0.09) and 0.92 (±0.09) and half recovery rates of 19.17 seconds (±6.87 seconds) and 21.10 seconds (±5.57 seconds) for stiff and solid dsLBs, respectively. Fig. 3D, E). These results show that PDMS crosslinking can be used to produce dsLBs with different degrees of stiffness, with core stiffness adjustable over three orders of magnitude. Remarkably, the dsLBs are able to retain their membrane fluidity during this process.

[0088] For the development of technologies for expanding synthetic T cells, the ability to effectively separate the expanded T cells from the aAPCs is a crucial practical aspect. This is particularly important for the safe reintroduction of T cells into patients.

[0089] A common method to achieve this separation is magnetic isolation in bead-based technologies. To seamlessly integrate dsLB into such established workflows, magnetic iron oxide nanoparticles (NPs) with a diameter of 6.6 nm (± 1.4 nm, n=10 particles, nuclei measured by electron microscopy) were synthesized. Fig. 16A). To incorporate these NPs into the hydrophobic dsLB core, their surface was functionalized with poly(dimethylsiloxane) using an ω-(carboxydecyl)-C10 anchor. Bright-field microscopy confirmed the successful integration of the NPs into the dsLB core, which could then be magnetically separated from the solution ( Fig. 16B). Our results showed that 70 mg / mL of NPs are sufficient for effective separation of dsLB from the solution ( Fig. 16C). 2.3 Activation and expansion of T cells by droplet-based lipid bilayers

[0090] To assess the potential of dsLBs to activate T cells for expansion, dsLBs conjugated with anti-CD3 and anti-CD28 (in a 1:4 ratio) were incubated with human primary CD8+ T cells isolated from platelet apheresis kits from healthy volunteer donors by negative selection (Knörck, A. et al. Quantity, quality, and functionality of peripheral blood cells derived from residual blood of different apheresis kits. Transfusion 58, 1516-1526, doi:https: / / doi.org / 10.1111 / trf.14616 (2018)). After a short incubation period of only 4 hours, frequent contact between the T cells and the dsLBs was observed, resulting in polarized T-cell morphology ( Fig. 12A). Furthermore, by using fluorescently labeled anti-CD3 antibodies immobilized on the dsLB surface, it was found that the T cells preferentially settled at sites of increased CD3 aggregation, suggesting antibody recruitment and IS formation ( Fig. 12B). Furthermore, after an expansion phase of 4 days, the formation of larger cell aggregates was observed, similar to T-cell cultures expanded using the Dynabead technology based on solid beads ( Fig. 12C). These results suggest that dsLBs have the potential to activate T cells through antibody binding and IS formation, leading to expansion.

[0091] The expansion of CD8+ T cells was also quantified when activated with dsLBs of varying stiffness and compared to expansions achieved with anti-CD3 / CD28 DynaBeads. The results showed that dsLBs were only slightly less effective than DynaBeads, exhibiting a 34–28% lower expansion rate ( Fig. 12D). Surprisingly, the stiffness of the dsLBs had no significant influence on the expansion rates. This observation is consistent with flow cytometric analyses of the activation marker CD25, which showed similar CD25+ populations in T cells expanded with dsLBs of different stiffness ( Fig. 12E and Gating Strategy in Fig. 17).

[0092] A crucial parameter for T-cell expansion for immunotherapy, besides their phenotype (e.g., memory, effector, regulatory), is their susceptibility to immunosuppression in the cancer microenvironment. To assess the susceptibility of dsLB-expanded T cells to immunosuppression, the expression of the immune checkpoint receptor PD1, which is typically upregulated by ex vivo T-cell expansion technologies, was investigated (Sudarsanam, H., Buhmann, R. & Henschler, R. Influence of Culture Conditions on Ex Vivo Expansion of T Lymphocytes and Their Function for Therapy: Current Insights and Open Questions. Front Bioeng Biotechnol 10, 886637 (2022). https: / / doi.org:10.3389 / fbioe.2022.886637). Remarkably, a significantly lower PD1 expression was found compared to DynaBeads-expanded cultures ( Fig. 13A). Again, no significant difference was found between dsLBs with different stiffnesses. Based on this PD1low A more detailed analysis of the T-cell phenotype following dsLB expansion was performed to uncover the underlying functional properties. Multiplex quantification of 12 key cytokines secreted by different T-cell subsets was carried out ( Fig. 13B). Compared to DynaBeads, dsLB-expanded CD8+ T cells secreted significantly lower amounts of effector-associated cytokines such as TNFα, FasL, and granzyme. Interestingly, dsLB-expanded CD8+ T cells produced higher levels of regulatory and suppressive cytokines, particularly IL-4 and IL-10, compared to DynaBead-expanded cells. DsLB-expanded CD8+ T cells also secreted substantial amounts of the pro-inflammatory cytokine IL-17 and the regulatory cytokine INFγ, albeit to a lesser extent than DynaBead-expanded T cells. The secretion of these cytokines, normally associated with regulatory and helper T cells, was unexpected in CD8+ cells and was not affected by dsLB stiffness or ligand density. This finding was also not due to CD4 contamination. + T cells through the CD8-specific negative selection process ( Fig. 18). Although IL-4 and IL-10 producing CD8+ T cells with a regulatory (Treg) or suppressive phenotype have already been described (Schwaiger, S., Wolf, AM, Robatscher, P., Jenewein, B. & Grubeck-Loebenstein, B. IL-4-producing CD8+ T cells with a CD62L++(bright) phenotype accumulate in a subgroup of older adults and are associated with the maintenance of intact humoral immunity in old age. J Immunol 170, 613-619 (2003). https: / / doi.org:10.4049 / jimmunol.170.1.613; Noble, A., Giorgini, A. & Leggat, JA. Cytokine-induced IL-10-secreting CD8 T cells represent a phenotypically distinct suppressor T-cell lineage. Blood 107, 4475-4483). (2006). https: / / doi.org:10.1182 / blood-2005-10-3994), their understanding remains limited, primarily due to inadequate expansion and isolation techniques and a lack of specific markers for their identification.To further characterize the phenotype of dsLB-expanded T cells, the expression of markers commonly associated with CD8+ Treg and memory differentiation was measured (. Fig. 13C,D). Comparison of DynaBead and soft dsLB-activated T cells showed that both expansion approaches resulted in only a small increase in Foxp3. + and CD103 + expressing T cells, markers recently associated with CD8+ Tregs were found (Mishra, S., Srinivasan, S., Ma, C. & Zhang, N. CD8+ Regulatory T Cell - A Mystery to Be Revealed. Frontiers in Immunology 12 (2021). https: / / doi.org:10.3389 / fimmu.2021.708874). The expression of CD27, specifically in activated CD95 + However, the expression of CCR7 and CD62L was varied in dsLB-expanded cultures compared to DynaBeads.

[0093] This suggests a differentiation from a central to an effector-memory-like phenotype, with a low proportion of highlighted cells, as measured by CD57 staining. Overall, the data indicate that dsLBs preferentially expand a differentiated T-cell subset with central to effector-like memory marker expression and IL-4 / IL-10-based regulatory and suppressive properties that are less susceptible to checkpoint inhibition (reduced PD-1 expression). All of these are favorable characteristics for applications in adaptive cell therapy. A three-dimensional parameter space for the expansion of T-cells

[0094] It was noted that our results are somewhat inconsistent with previous studies showing that variations in aAPC stiffness significantly influence T-cell activation. However, no such pronounced effects were observed in the present dsLB-based system. This discrepancy could be due to the use of dispersed aAPCs with lateral ligand mobility, which facilitates molecular self-assembly at the IS level. Lateral ligand mobility may mask or attenuate stiffness-related mechanosignaling in T cells, as has been previously observed in other systems (Lippert, AH et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal (2021) 120, 35-45, doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021).To investigate this hypothesis, the expansion of T cells was studied using non-deformable silicon dioxide microspheres with a stiffness orders of magnitude higher (70 GPa) coated with a lipid bilayer (memBeads) that is identical in composition to the dsLBs present (. Fig. 14A), and which are functionalized with the same anti-CD3 / CD28 antibodies via protein G coupling at the same density. These beads are comparable to DynaBeads in size (4.5 µm) and stiffness, but additionally feature laterally movable ligands. No significant difference was found in their ability to activate T cells, as determined by CD25+ cell quantification, compared to the approximately one order of magnitude softer dsLBs ( Fig. 14B). Interestingly, the memBeads also expanded T cells with a PD1 low expressing phenotype ( Fig. 19) It is noteworthy that relevant parameters such as antibody clones, antibody densities, aAPC:T cell ratio, and membrane composition were similar between the two aAPC substrates. This supports the hypothesis that an aAPC membrane can mask the effects caused by aAPC stiffness and is the main reason for PD1. low expressing phenotype.

[0095] Given the well-established importance of mechanosignaling in T-cell activation, which underlies the three main biochemical signals (TCR activation, co-stimulatory ligands, cytokines), it was investigated whether ligand mobility might also mask effects resulting from variations at the level of biochemical signaling. Changing the anti-CD3 antibody (signal 1) from a UCHT1 clone to a lower-affinity SK7 clone led to a decrease in CD25+ cells, although the expansion rates of eventual T cells remained largely unaffected. Fig. 14C, D). A similar result was observed when the ratio of the stimulating antibodies Signal 1 and 2 was changed from 1:4 to 1:1, resulting in a slightly smaller CD25+ population with minimal effects on expansion rates. No significant differences in stiffness were observed in any of these stimulation scenarios, including when using memBeads. Furthermore, replacing the Signal 2 stimulant anti-CD28 with the integrin ligand ICAM1 resulted in a significant reduction in CD25+ cells and minor variations in T-cell expansion. Changes at the level of Signal 3, in this case IL-2, similarly affected T-cell activation and expansion. Reducing the IL-2 concentration from 50 U / mL to 12.5 U / mL resulted in a significant reduction in both T-cell expansion and the number of CD25+ cells.Overall, these results suggest that while the lateral mobility of ligands on aAPCs can mask stiffness-related effects, the dominance of biochemical signaling over biophysical cues remains intact.

[0096] It was assumed that biochemical signaling by agonistic CD3 antibodies was so potent that it could override any subtle effects mediated by stiffness variations. This might also be the case if the signal intensity of the TCR trigger was reduced by using the SK7 anti-CD3 clone, a lower-affinity but still effective anti-CD3 agonist. Consequently, it was hypothesized that using an even lower biochemical signal intensity at the level of signals I and II might allow latent, stiffness-related effects, masked by the lipid membrane, to become more apparent. This would suggest that biochemical signaling, substrate stiffness, and ligand mobility are three separate but synergistic factors contributing to the nuanced process of T-cell activation.To further investigate this potential interplay, a lower signal strength TCR stimulation configuration was developed that is also relevant for immunotherapy by using a bispecific T-cell engager (TCE). Fig. 14E). dsLBs were formed that present the tumor-associated antigen (TAA) Her2, and the surface protein density was reduced to 100 (low) and 400 (high) molecules / µm. 2 set using MESF beads ( Fig. 20). In conjunction with an ICAM1-based signal 2 and 50 U / mL IL-2 signal 3, a bispecific TCE targeting Her2 and CD3 was added to dsLBs in CD8+ T-cell cocultures to initiate signal 1. Here, dsLBs functioned as cancer-target cell mimetics in a TCE-mediated in vitro mimicked immunotherapy. The size of the Her2-TCE-CD3 complex is approximately twice that of the anti-CD3 / TCR complex, and the kinetic segregation model (Davis, SJ & van der Merwe, PA. The kinetic-segregation model: TCR triggering and beyond. Nature Immunology 7, 803-809, (2006) doi:10.1038 / ni1369.) as well as earlier data comparing the efficacy of IS formation between anti-CD3 agonists and bispecific T-TCEs (Staufer, O. et al. Solution structure and synaptic analyses reveal molecular mechanisms of bispecific T cell engagers. bioRxiv, 2022.2006.2015.496334 (2023). https: / / doi.org:10.1101 / 2022.06.15.496334) indicate a lower stimulation intensity of these dsLBs on T cells compared to anti-CD3 / CD28 dsLBs. Confocal microscopy showed successful adhesive contact formation between Her2-dsLBs and T cells. Furthermore, relocalization of AlexaFluor488-labeled Her2 at the cell interface was observed, indicating successful TCE coupling and IS formation (. Fig. 14F). Subsequently, the IC50 of the TCE and its maximum activation potential (as indicated by CD25 upregulation) were quantified for two ligand densities on soft and rigid dsLBs ( Fig. 14G). Stiff dsLBs were not included in the analysis because their stiffness exceeds that of cancer cells by several orders of magnitude and is therefore most likely not relevant for TCE therapies (Xu, W. et al. Cell Stiffness Is a Biomarker of the Metastatic Potential of Ovarian Cancer Cells. PLoS one 7, e46609 (2012). https: / / doi.org:10.1371 / journal.pone.0046609; Han, YL et al. Cell swelling, softening and invasion in a three-dimensional breast cancer model. Nat Phys 16, 101-108 (2020). https: / / doi.org:10.1038 / s41567-019-0680-8). The IC50 values ​​differed significantly by two orders of magnitude between soft and stiff dsLBs. The maximum activation efficiency was also influenced by the substrate stiffness and the Her2 density.These results confirm that substrate stiffness can influence T-cell activation even when ligands are laterally mobile, but especially when biochemical signal strength is reduced. This suggests a trigger hierarchy for T-cell activation, in which biochemical signals provide the strongest cue, followed by biophysical cues related to lateral ligand mobility, and finally by the stiffness of the aAPC itself (biochemical signal strength > ligand mobility > substrate stiffness).

[0097] Crucially, PD1 expression was also investigated in this TCE setup, and it was found that the PD1-expressing population remained significantly lower at both dsLB severity levels compared to DynaBead-positive controls ( Fig. 14H). This result suggests that it is primarily the introduction of laterally mobile ligands that promotes a T-cell phenotype with lower susceptibility to immunosuppression.

[0098] This invention introduces a novel aAPC technology inspired by the principles of synthetic bottom-up biology, designed to mimic the biophysical properties of APC membranes. The dsLB technology combines lateral membrane mobility and stiffness modulation to integrate an additional biophysical factor capable of tuning T-cell-aAPC interactions. Previous studies have focused on the mechanobiological component of T-cell activation (Harrison, DL, Fang, Y. & Huang, J. T-Cell Mechanobiology: Force Sensation, Potentiation, and Translation. Frontiers in Physics 7, (2019) doi:10.3389 / fphy.2019.00045.). However, the significance of lateral ligand mobility in this context has not yet been fully explored due to a lack of suitable methods that consider both aspects. While solid and planar lipid bilayers are used to investigate the formation dynamics of IS (Hsu, C.-J. et al.Ligand Mobility Modulates Immunological Synapse Formation and T Cell Activation. PLoS ONE 7, e32398, (2012). doi:10.1371 / journal.pone.0032398.) Investigating the individual contributions and functional effects of stiffness and mobility factors on the expansion of therapeutic T cells using suitable dispersed expansion systems was a challenge. The results confirm that mechanical factors, together with lateral ligand mobility, provide a biophysical context for receptor-ligand-mediated T cell activation. They also demonstrate how T cell differentiation can be guided by such biophysical cues and underscore that lateral ligand mobility is an important feature for the development of effective aAPC technologies.Furthermore, observations that dsLB produces a smaller population of PD1-expressing T cells after expansion suggest that membrane mobility may be crucial for controlling the differentiation process during ex vivo T-cell expansion for therapeutic approaches. Since this has also been observed with silicon dioxide bead membranes and in our bispecific T-cell engager system, lateral ligand mobility, rather than stiffness, may be critical for this effect.

[0099] Most importantly, it was found that dsLB has a CD8 + The T-cell phenotype expands, secreting the anti-inflammatory and regulatory cytokines IL-4 and IL-10. These are typically associated with regulatory CD4 cells. + T-cells have been associated with them, but have already been repeatedly identified as CD8 +T cells have been described, including their potential role in immunotherapy (Flippe, L., Bézie, S., Anegon, I. & Guillonneau, C. Future prospects for CD8(+) regulatory T cells in immune tolerance. Immunol Rev 292, 209-224 (2019). https: / / doi.org:10.1111 / imr.12812). Immunotherapeutic approaches targeting regulatory CD8 + The development of T cells based on this phenotype is currently hampered by the lack of suitable ex vivo expansion technologies that favor this phenotype over cytolytic CD8+ T cells as the effector. Although a comprehensive functional analysis of dsLB-expanded T cells is still pending to evaluate their suppressive and cytolytic potential in a therapeutic context, their surface marker expression profile suggests a differentiation stage closely related to effector and central memory T cells.

[0100] The dsLB technology was compared to a widely used solid bead-based method (DynaBeads), a globally established technology. Although the expansion rates achieved with dsLB were consistently lower, dsLB represents a potential complement to this approach and expands the repertoire for therapeutic T-cell expansion due to the lower proportion of PD1-expressing T cells and their marked differences in IL-4 and IL-10 secretion.

[0101] It is noteworthy that the observed effects on T-cell activation and differentiation were not mediated by biophysical cues related to stiffness, but most likely by the presentation of stimulating ligands on the laterally mobile membrane. In this context, the central role of receptor organization at the IS for T-cell activation is generally accepted in the field (Dustin, ML. The immunological synapse. Cancer Immunol Res 2, 1023-1033, (2014) doi:10.1158 / 2326-6066.Cir-14-0161). DsLBs offer a first α-APC technology for the generation of dispersed, low-stiffness, and scalable particles with laterally mobile ligands.

[0102] Since the stiffness-related effects were not as pronounced as observed in other studies, the present systematic analysis of the effects of substrate stiffness and the signal strength of signals I, II and III on T-cell activation suggests a three-dimensional physicochemical space of T-cell activation ( Fig. 15). In this space, lateral ligand mobility appears to be the main switch for PD1 expression.

[0103] DsLBs have other interesting applications in biomaterials science and synthetic biology. While previous studies have described the formation of PDMS droplets for sensor purposes, few have demonstrated their potential for use in biomedicine (Lambert, LH et al. Improving T Cell Expansion with a Soft Touch. Nano Lett 17, 821-826, (2017) doi:10.1021 / acs.nanolett.6b04071; Park, S., Song, S. & Yoon, S.-H. Ultrasonication-induced and diluent-assisted suspension polymerization for size-controllable synthesis of polydimethylsiloxane droplets. Colloids and Surfaces A: Physicochemical and Engineering Aspects 644, 128827, (2022). doi:https: / / doi.org / 10.1016 / j.colsurfa.2022.128827; Pan, C. et al. Silver-Coated Poly(dimethylsiloxane) Beads for Soft, Stretchable, and Thermally Stable Conductive Elastomer Composites. ACS Applied Materials & Interfaces 11, 42561-42570, (2019). doi:10.1021 / acsami.9b13266; Jiang, K., Thomas, PC, Forry, SP, DeVoe, DL& Raghavan, SR. Microfluidic synthesis of monodisperse PDMS microbeads as discrete oxygen sensors. Soft Matter 8, 923-926, (2012). doi:10.1039 / C2SM06685H.). In the present study, an adjustable stiffness of these PDMS droplets was achieved by employing a differentiated curing process. Furthermore, a novel approach for functionalizing and stabilizing the dispersed droplets in solution was developed, utilizing ionic surfactants to generate a charge-dependent lipid bilayer on the droplets. This approach, along with magnetic nanoparticle-based separation, has significant implications for scaling up the technology and increasing its applicability. As demonstrated by electron microscopic observations and stability experiments, it also imparts stability to the emulsions to prevent droplet aggregation or segregation.By opening a new route for the production of lipid bilayers on droplets, the applications of dsLBs in biomedicine and beyond have been expanded. Potential improvements to the technology include the implementation of a membrane extrusion-based or microfluidically assisted emulsification process to increase the monodispersity of the dsLBs (Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203.). Furthermore, additional coregulatory ligands, such as CD58 or membrane-immobilized IL-2, could be integrated into a dsLB system to provide additional means for controlling the differentiation process (Capera-Aragones, J. et al. Understanding the spatial and functional link between the IL-2R and TCR at the immunological synapse.Biophysical Journal 122, 510a (2023). https: / / doi.org:https: / / doi.org / 10.1016 / j.bpj.2022.11.2717). DsLBs could be used in areas such as drug delivery, biosensors, and tissue engineering, where they could represent a more versatile and effective alternative to existing biomimetic synthetic cell techniques. In summary, by harnessing the power of bottom-up synthetic cell synthesis as cellular bionics of aAPCs, a novel approach to T-cell activation has been developed that can mimic the key biochemical and biophysical properties of APCs. Materials and methods for DsLB production

[0104] The DsLBs were after the in Fig.The strategy described in section 1 was used. Approximately 100 mg of PDMS (Sylgard 184, Dow Corning, USA) was mixed with 880 µl of PBS containing SDS (1 mM) and pre-emulsified manually by resuspension. The solution was then placed in an ultrasonic bath for 2 minutes at room temperature to produce a dispersed O / W emulsion. Subsequently, MgCl₂ (Sigma-Aldrich, Germany) was added to a final concentration of 20 mM (unless otherwise specified), along with 100 µl of a 6 mM (final lipid concentration) SUV solution to form dsLBs. SUVs consisting of 20 mol% EggPG, 5 mol% PE-DGS-NTA(Ni) 2+), 1 mol% LissRhodamin B-PE or Atto488-PE and 74 mol% EggPC (all Avanti Polar Lipids, USA) were prepared by extrusion as previously described (Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203 (2021).). (SUVs consisting of different lipid compositions were prepared by dissolving the desired lipids in pure chloroform, mixing them to the desired composition and concentration, and drying them under a gentle stream of nitrogen. To remove traces of the solvent, the lipids were held under vacuum in a desiccator for approximately one hour. The dried lipids were then resuspended by adding the desired aqueous buffer (e.g., PBS), followed by one hour of vortexing.)SUV size was homogenized by extruding the solution 7–11 times through a 50 nm pore size polycarbonate filter (Whatman, Germany) using an extruder (Avanti Polar Lipids, USA). Solutions containing SUVs were stored at 4°C. The dsLB solution was incubated at room temperature, protected from light, for 5 minutes and then centrifuged at 8600 x g for 30 seconds. The supernatant was discarded, and the dsLB pellet was resuspended in 1 mL of PBS for a further centrifugation step using the parameters mentioned above. The dsLB solution was resuspended in 1 mL of PBS and stored at 4°C, protected from light, until further use. For the control experiment, 1 mM Triton X-100 (ACROS Organics) was used instead of SDS.Where indicated, dsLBs were also formed from other silicone oils, in particular AR20 (Sigma Aldrich, Germany), AP100 (Sigma Aldrich, Germany) and highly thermostable SO (Merck, Germany).

[0105] For protein functionalization, the total amount of accessible dsLB membrane in a dsLB preparation batch was measured by calibration against an SUV dilution series of known concentration. For this purpose, an SUV dilution (the same batch used for dsLB preparation) was prepared in a 96-well plate. The fluorescence intensity (originating from Rhodamine B) in each dilution was measured using a TECAN Spark (Tecan Group, Switzerland) plate reader controlled by TECAN SparkControl software with integrated gain optimization and an excitation / emission setting of 537 / 582 nm to generate a fluorescence-concentration calibration curve. The total lipid concentration of a dsLB solution was determined by measuring the fluorescence intensity across the SUV dilution series and referencing the calibration curve.The concentration of total accessible DGS-NTA (Ni) was derived from the final lipid concentration. 2+ ) derived. Protein G (Sigma Aldrich, Germany) was added in the molar ratio indicated in the figure captions and allowed to bind to the NTA(Ni). 2+)-units were incubated at 4 °C for 60 minutes, protected from light. The dsLBs were then washed by centrifugation before the addition of IgG antibodies against human CD3 (UCHT1, Invitrogen), CD28 (CD28.2, Invitrogen), CD8 (SK1, BioLegend), or CD3-AlexaFluor488 (HIT3a, BioLegend) at a ratio of 1:1.5 protein G to IgG. The ratio of CD3 and CD28 antibodies was adjusted to 1:4. The IgGs were incubated at 4 °C for 30 minutes, protected from light, to bind to the protein G on the dsLBs. The remaining IgGs were then washed off by centrifugation, and the final dsLB lipid concentration was quantified using a plate reader, as described above. Networking of dsLB oil droplets

[0106] To achieve adjustable stiffness of the dsLBs, the PDMS oil droplets were crosslinked. For this purpose, the PDMS oil was thoroughly mixed with silicone elastomer curing agent (Sylgard 184, Dow Corning USA) at a ratio of 10:1 (stiff dsLBs) and 50:1 (firm dsLBs) and then resuspended in 880 µL of PBS containing 1 mM SDS. This oil-in-water emulsion was treated ultrasonically for 2 minutes and then cured overnight at 60 °C in a water bath. dsLBs were then fabricated from the cured droplets as described above. Environmental scanning electron microscopy

[0107] For ESEM-based analysis, dsLBs from a standard batch (1 mM SDS, 20 mM MgCl₂) were pelletized by centrifugation and resuspended in 100 µl of PBS. A 5 µl droplet of this solution was placed on an FEI Quanta 400 FEG equipped with an ESEM pelleting stage at 3 °C and transferred to the microscope chamber. A water vapor pressure of 750 Pa was chosen to start with fully humid conditions. The water vapor pressure was then gradually reduced, causing the water droplet to shrink due to water evaporation. Individual dsLBs were visualized at different water vapor pressures at 3 °C. CryoTEM

[0108] For cryoTEM, a 2 µL drop of the sample solution was applied to a perforated carbon film (Plano, S147-4), dabbed for 2 s, and vitrified in liquid ethane in a Gatan CP3 immersion freezer at -165 °C. The resulting cryoTEM sample was transferred to a Gatan cryoTEM sample holder, model 914, and imaged using bright-field transmission electron microscopy (JEOL JEM 2100 LaB6) at an acceleration voltage of 200 kV under low-dose conditions. A Gatan Orius SC1000 CCD camera was used for image acquisition (exposure time 2 s). Dynamic light scattering

[0109] The zeta potentials of dsLBs and SUVs were measured using a Malvern Zetasizer Nano ZS system. The equilibration time was set to 30 s at 25 °C, followed by three repeat measurements for each sample at a scattering angle of 173°, using the built-in automatic sequence numbering. The refractive index of the material was set to 1.460, the absorbance to 0.010, and the solvent properties to η = 0.8882, n = 1.33, and ε = 79.0. All measurements were performed in 1X phosphate-buffered saline (pH 7.4). NBD provision

[0110] For the evaluation of the lipid layer configuration by NBD tests, a modified, previously published protocol was used (Ohmann, A. et al. A synthetic enzyme built from DNA flips 10 7 lipids per second in biological membranes. Nature communications 9, 2426, (2018) doi:10.1038 / s41467-018-04821-5.).

[0111] DsLBs were prepared from 20 mol% EggPG, 1 mol% PE-DGS- NTA(Ni 2+), 1 mol% 16:0-12:0 NBD PE and 78 mol% EggPC SUVs were formed. NBD fluorescence was measured using a Tecan Spark plate reader controlled by TECAN SparkControl software with integrated gain optimization and an excitation / emission setting of 468 / 537 nm. A dithionite dilution series (Sigma Aldrich, Germany) between 0.2 and 50 mM was prepared in a 96-well plate, and dsLBs were added to a final concentration of 60 µM total lipids (assuming all SUVs are incorporated into the dsLB membrane). The fluorescence intensity and the amount of NDB fluorophores were correlated using an SUV dilution series, and the nonlinear fluorescence intensity of dsLBs was corrected using a serial dilution of dsLBs. Both calibrations were prepared and measured on the same plate to generate a calibration curve of fluorescence intensity to the NDB fluorophores.The number of fluorophores was normalized to the untreated dithionite control. Since dithionate is a strong acid, all experimental steps were carried out in PBS + 100 mM HEPES. Confocal microscopy

[0112] Confocal microscopy was performed using an LSM 880 laser scanning microscope (Carl Zeiss AG). Images were acquired with a 20× (Plan-Apochromat 20x / 0.8 M27, Carl Zeiss AG, Germany) and a 63× immersion oil objective (Plan-Apochromat 63x / 1.4 oil DIC M27, Carl Zeiss AG, Germany) and with laser lines of 488 and 543 nm. The images were evaluated using ImageJ software (NIH). For imaging, dsLBs were resolved at a ratio of 1:20 or 1:40 in Nunc LabTeK glass bottom chambers with 8 wells and filled with 200 µL of PBS. Analysis of size distribution

[0113] For the analysis of the dsLB size distribution, dsLBs were imaged using confocal microscopy with an LSM 880 (Carl Zeiss AG) equipped with a 20x objective and an excitation / emission setting of 458 / 543 nm. Z-stacks were acquired across the entire dsLB height, from which maximum z-projections were generated. Particle sizes were measured using ImageJ software (NIH, USA) by global segmentation based on intensity thresholds, separation of particles by watersheds, and automatic particle detection. Restoration of fluorescence after photobleaching analysis

[0114] FRAP experiments were performed using an LSM 880 laser scanning microscope (Carl Zeiss AG) with a 63x objective. Bleaching areas in the equatorial plane of the dsLBs with a minimum radius of 5 µm were selected and bleached with a 100% laser intensity of 488 nm. At least 3 images were acquired before bleaching and 27 images after bleaching. Mean time profiles of the fluorescence intensity for the bleached area, the entire dsLB, and the background fluorescence were measured using ImageJ software (NIH). Background and intensity profile normalization, data fitting, mobile fraction calculation, and t-value calculation were performed. 1 / 2 Maximum measurements were performed using the web-based tool easyFRAP (Zoi Lygerou, University of Patras). MESF bead assay

[0115] DsLBs were combined and the concentration of DGS-NTA(Ni) 2+), as described above. To account for possible non-specific binding of protein G or antibodies, additional dsLBs without DGS-NTA(Ni) were calculated. 2+ ) produced. To functionalize the dsLBs, the samples were mixed with different amounts of protein G. The following molar ratios of protein G to DGS-NTA(Ni) were obtained. 2+ The following ratios were used: 1:8, 1:4, 1:2, 1:1 and 2:1. For the DGS-NTA(Ni 2+The same amount of protein G as in the 1:2 mixtures was used for the dsLBs (dissociated solid bases). All samples were incubated at 4°C for 45 minutes and then washed by centrifugation. Alexa Fluor 488-labeled anti-CD8 (SK1, BioLegend, UK) was added to each sample, with the antibody concentration adjusted to 150% of the respective original protein G concentration. The samples were protected from light, incubated at 4°C for 45 minutes, and then washed by centrifugation. A composition of Quantum Alexa Fluor 488 MESF beads (Bangs Laboratories Inc., USA) was prepared according to the supplier's instructions. The DsLBs were premixed 1:1 with the bead composition and dissolved 1:10 in Nunc LabTeK 8-well glass-bottom chambers filled with 200 µl of PBS. Imaging was performed with an LSM 880 (Carl Zeiss AG), using a 20x lens and 488nm excitation.The software ImageJ (NIH) was used for particle separation in the water, automatic particle detection, and measurement of the mean fluorescence intensity of individual particles. The beads were identified based on the absence of rhodamine B fluorescence signals and their uniform size and used to generate a calibration curve to correlate the fluorescence signal with the MESF. The mean number of antibodies on the dsLBs was calculated and used to generate another calibration curve for the amount of protein G / antibody used and the number of antibodies bound to the particle surface. Real-time deformation cytometry

[0116] The elastic modulus of uncured dsLBs was measured using an AcCellerator (ZellMechanik Dresden, Germany), a commercial real-time deformation cytometer (RT-DC). RT-DC analysis involved introducing a flow into a microfluidic chip using syringe pumps. The dsLBs were deformed by hydrodynamic stresses as they passed through a narrow constriction larger than their diameter. Imaging was performed with an inverted microscope (Axiovert 200M, ZEISS, Germany), and images were captured at the end of the microfluidic constriction using a CMOS camera (Mikrotron, Germany). ShapeIn software (Zellmechanik Dresden, Germany) was used to control the camera and syringe pump.The software ShapeOut 2 (Cell Mechanics Dresden, Germany) was used to analyze the acquired images in real time, enabling the calculation of a contour and a projected area for each dsLB. To exclude dsLB duplicates or debris, the surface boundaries were set to 80 and 250 µm. 2 and the deformation limits were set to 0.003 and 0.008. For the measurements of dsLBs, the samples were diluted according to a McFarland standard of 6 MFS with a viscosity-matched CellCarrier buffer (Zellmechanik Dresden, Germany), thereby increasing the stresses acting on the dsLBs in the channel. The measurements were performed in microfluidic channels with a width of 20 µm at a flow rate of 0.025 µL / s. Synthesis of iron oxide nanoparticles

[0117] Iron oxide nanoparticles (NPs) were synthesized as described (Xu, Z., Shen, C., Hou, Y., Gao, H. & Sun, S. Oleylamine as Both Reducing Agent and Stabilizer in a Facile Synthesis of Magnetite Nanoparticles. Chemistry of Materials 21, 1778-1780 (2009). https: / / doi.org:10.1021 / cm802978z). Specifically, 6 mmol of iron(III) acetylacetonate (Sigma Aldrich, 97%) was dissolved in 30 mL of dibenzyl ether (Sigma Aldrich, ≥ 98.0% GC) and 30 mL of oleylamine (Acros Organics, 80-90%). The solution was heated to 110°C to 130°C for 1 hour under a continuous nitrogen flow. The solution was then heated to 300°C under a constant nitrogen atmosphere at a heating rate between 15°C / min and 20°C / min and held at 300°C for 1 hour. The dispersed NPs were cooled to room temperature and washed three times with a 50 / 50 mixture of isopropanol and acetone. Finally, the NPs were dispersed in toluene at a concentration of 40 mg / mL.Ligand exchange to poly(dimethylsiloxane), ω-(carboxy decyl)-terminated (Polymer Source, P18578-DMS-C. 10 -COOH) was carried out as follows: 20 mL of the 40 mg / mL iron NP dispersion were heated to 80°C with stirring. After reaching the temperature, 1 mL of the ligand was added to the dispersion. The mixture was stirred at 80°C for 24 hours. The dispersion was then cooled, washed three times with ethanol, and finally dispersed in 20 mL of hexane (Sigma Aldrich ≥95%). Compression rheology

[0118] To measure the modulus of elasticity of PDMS bulk material for stiff and solid dsLBs, PDMS was cured in a 24-well plate overnight at 60 °C in a drying oven at ratios of 10:1 and 50:1 with crosslinker. The cured PDMS mass was punched to a diameter of 10 mm, resulting in PDMS blocks with a 1:1 diameter-to-height ratio. The moduli of elasticity of the punched PDMS bulk cylinders were measured by compression rheology using the Z100 (ZwickRoell, Germany), equipped with a 200 N load cell and a 40 mm compression ram. The test speed for PDMS volume compression was 1 mm / min, while the preload was 0.10 N for stiff and 0.05 N for solid PDMS bulk material. Nanoindentation analysis

[0119] To confirm the elastic moduli for PDMS bulk material, the samples were measured using the Pavone nanoindenter (OPTICS11 life, Netherlands). For this purpose, the bottom of 48-well plates was coated with uncured, degassed PDMS, which was crosslinked with curing agent at a ratio of 10:1. To stabilize the dsLBs, they were prepared according to the protocol for crosslinking dsLB oil cores. 50 µL of the dsLB suspension was mixed with 950 µL of PBS and injected into the coated wells. The samples were cured overnight at 60 °C in a drying oven. BSA was added to the PBS at a final concentration of 1% and incubated for 10 minutes at room temperature. The dsLBs were investigated using a cantilever with a tip diameter of 3 µm and a stiffness of 0.53 N / m at an indentation of 800 nm. The data were processed and analyzed using Inbuild software and a Herzian fit. Isolation and culture of T cells

[0120] Human primary CD8 + were routinely cultured in RPMI 1640 w / L-glutamine (VWR, Germany) medium supplemented with 10% fetal bovine serum (Gibco, Germany), 1% penicillin / streptomycin (Gibco, Germany), 1% non-essential amino acids (Biowest), and 50 mM HEPES (Sigma Aldrich, Germany) in cell culture flasks at 37 °C, 5% CO2, and 100% relative humidity. For the cultivation and expansion of CD8 + For all functional tests, 50 U / mL of recombinant human IL-2 (STEMCELL technologies, Germany) was added to the culture medium. Primary human CD8 + were obtained from leukapheresis reduction system (LRS) chambers of unidentified, voluntary healthy blood donors using commercially available negative selection kits (RosetteSep Human CD8). +T-cell enrichment cocktail (STEMCELL technologies, Germany) was isolated according to the manufacturer's instructions. Blood samples were provided by the Institute for Clinical Haemostaseology and Transfusion Medicine of Saarland University Medical Center, in accordance with Ethics Agreement No. 34 / 23 (Ethics Committee of the Saarland Medical Association). The cells were either used for the experiment within two days or cryopreserved. T-cell activation and expansion using dsLBs and flow cytometric analysis

[0121] The activation and expansion of primary human CD8 +T-cell expansion was performed in 48-well plate formats by incubating 240,000 T cells with dsLBs in an approximate 1:2 ratio in a total volume of 1.3 mL. A final concentration of 50 U / mL of recombinant human IL-2 was added for all tested conditions. Expansion was carried out for 9 days at 37 °C, 5% CO2, and 100% relative humidity in fully supplemented cell culture medium. The dsLBs were conjugated with the specified antibody densities immediately prior to the experiment. For control experiments, a final concentration of soluble anti-CD3 (UCHT1, Invitrogen) and anti-CD28 (CD28.2, BioLegend) antibodies or plain dsLBs without conjugated antibodies were added to the cells. Dynabeads human T-Activator CD3 / CD28 beads (Gibco, Germany) were used as positive controls according to the manufacturer's recommendations. After the expansion phase, the cell solution was resuspended and transferred to a 96-well, round-bottomed plate.The cells were pelleted at 300 g for 5 minutes, the supernatant was discarded, and the cell pellet was resuspended in PBS + 1% BSA with the staining antibodies (1:400) and incubated at room temperature for 30 minutes, protected from light. For surface marker staining, AlexaFluor488 or AlexaFluor647, FITC or PerCP / Cyanine 5.5 conjugated staining antibodies against CD25 (BC96, BioLegend, UK), PD-1 (NAT105, BioLegend, UK), CD8 (SK1, BioLegend, UK), CD4 (OKT4, BioLegend, UK), FOXP3 (259D, BioLegend, UK), CCR7 (G043H7, BioLegend, UK), CD62L (DREG-56, BioLegend, UK), CD103 (Ber-ACT8, BioLegend, UK), CD57 (HNK-1, BioLegend, UK), CD95 (DX2, BioLegend, UK) or CD122 (TU27, BioLegend, UK) were used. The cells were then washed by centrifugation and resuspension in PBS with 1% BSA to remove unbound antibodies.After a further centrifugation step, the cells were fixed in 2% PFA (Sigma Aldrich, Germany) for 30 minutes at room temperature, protected from light. The PFA solution was removed by centrifugation, and the cells were stored in PBS with 1% BSA at 4°C, protected from light, until further analysis. Surface marker staining was quantified using a Luminex Guava easyCyte flow cytometer with 642 nm and 488 nm laser lines. A minimum of 1000 cells was collected for each condition. The flow cytometric data were analyzed using FlowJo V.10 software (FlowJo LLC, USA). Cytokine analysis using flow cytometry

[0122] To gain a better understanding of the phenotypes of dsLB-expanded T cells, a LEGENDplex™ Human CD8 / NK Panel V02 (BioLegend, UK) was performed. For this purpose, primary human CD8 cells were used. +T cells were co-cultured with soft, firm, and rigid dsLBs, DynaBeads, and memBeads. These aAPCs were decorated with anti-CD3 (UCHT1 or SK7, Invitrogen) as the primary stimulating signal and anti-CD28 (CD28.2, BioLegend) as a co-stimulating signal in a 1:4 or 1:1 ratio with a final AB concentration of 200 molecules / µm (Tantalo, DGM et al. Understanding T cell phenotype for the design of effective chimeric antigen receptor T cell therapies. Journal for ImmunoTherapy of Cancer 9, e002555, (2021) doi:10.1136 / jitc-2021-002555). To account for co-stimulatory signals for T-cell expansion, CD28 was replaced by ICAM1 (Sino Biological, China) or IL-2 was reduced to 12.5 U / mL. The co-cultures were treated with 70,000 CD8 cells. +T cells were incubated at a 1:2 ratio with aAPC in 96-well plates for 7–9 days at 37 °C, 5% CO2, and 100% relative humidity in fully supplemented cell culture medium. After incubation, the cells were centrifuged, and 100 µL of the supernatant was used to perform the LEGENDplex™ assay. This was performed according to the manufacturer's instructions and measured using the FACSVerse flow cytometer. Cytokine concentrations were determined by analyzing the data using LEGENDplex™ compatible software from BioLegend (https: / / www.biolegend.com / en-us / immunoassays / legendplex / support / software). CD8 + T-cell expansion analysis

[0123] T cells from the previously described LEGENDplex™ assay were fixed with 2% PFA for 30 minutes at room temperature in a 96-well plate. After fixation, 2% BSA was added for 10 minutes before the cells were incubated for 1 hour with the nuclear staining fluorophore Hoechst 33342 (Thermo Fisher Scientific, Germany) at a final concentration of 1 µg / mL. The stained nuclei were imaged under UV excitation using a Leica DMI600 B (Leica Microsystems, Germany) and a 10x objective (HCX PL FLUOT AR 10x / 0.30, Leica Microsystems, Germany). The images were analyzed using ImageJ software (NIH, USA) by global intensity threshold segmentation, watershed particle separation, and automatic cell counting. Stimulation of T cells with dsLBs and a bispecific T-cell engager

[0124] CD8+ T cells were activated and expanded in 96 half-well plates by incubating 50,000 T cells with dsLBs in an approximate 1:2 ratio in a total volume of 180 µL of fully supplemented cell culture medium for 9 days. All conditions were further supplemented with recombinant human IL-2 at a concentration of 50 U / mL. Incubation conditions were 37°C, 5% CO2, and 100% relative humidity. The dsLBs were decorated with Her2 (Sino Biological, China) and ICAM1 (Sino Biological, China) using the same method previously described for protein G functionalization, with final Her2 protein densities of 100 molecules / µm² and 400 molecules / µm², respectively. A CD3 / Her2 bispecific T-cell engager (BSAB-002, Creative Biolabs, USA) was added to the medium in a six-step dilution series from 1 pM to 100,000 pM.After 4 to 5 days of incubation, the antibody-containing medium was replaced with fresh cell culture medium without additional TCE. Four control experiments were performed: T cells were incubated without dsLBs or with non-functionalized, pure dsLBs. T cells were co-cultured with dsLBs decorated with 100 Her2 molecules / µm (Tantalo, DGM et al. Understanding T cell phenotype for the design of effective chimeric antigen receptor T cell therapies. Journal for ImmunoTherapy of Cancer 9, e002555, (2021) doi:10.1136 / jitc-2021-002555.), while the cell culture medium was supplemented with a final concentration of 100,000 pM human serum IgG (I4506, Sigma Aldrich, Germany). Finally, Dynabeads human T-Activator CD3 / CD28 Beads (Gibco, Germany) served as a positive control and were used according to the manufacturer's recommendations.T cells were stained with AlexaFluor488- or AlexaFluor647-conjugated antibodies against CD25 (BC96, BioLegend, UK) and PD-1 (NAT105, BioLegend, UK) and fixed in 2% PFA (Sigma Aldrich, Germany) as described above. Surface marker staining was documented using Luminex Guava easyCyte and FACSVerse flow cytometers. At least 2000 cells were collected for each condition. Flow cytometric data were analyzed using FlowJo V.10 software (FlowJo LLC, USA). Post-processing of the data and statistical analysis

[0125] The diagrams were drawn using either GraphPad Prism 8 or Origin 2022 (OriginLab). The statistical analysis was performed, as indicated, using the integrated functions of GraphPad Prism 8. Some of the figures were created using Biorender. Cited literature 1) Waldman, A. D., Fritz, J. M. & Lenardo, M. J. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nature Reviews Immunology 20, 651-668, (2020) doi:10.1038 / s41577-020-0306-5. 2) Tantalo, D. G. M. et al. Understanding T cell phenotype for the design of effective chimeric antigen receptor T cell therapies. Journal for ImmunoTherapy of Cancer 9, e002555, (2021) doi:10.1136 / jitc-2021-002555. 3) Zhang, D. K. Y. et al. Enhancing CAR-T cell functionality in a patient-specific manner. Nature communications 14, 506, (2023) doi:10.1038 / s41467-023-36126-7. 4) Li, Y. & Kurlander, R. J. Comparison of anti-CD3 and anti-CD28-coated beads with soluble anti-CD3 for expanding human T cells: Differing impact on CD8 T cell phenotype and responsiveness to restimulation. Journal of Translational Medicine 8, 104, (2010) doi:10.1186 / 1479-5876-8-104. 5) Isser, A., Livingston, N. K. & Schneck, J. P. Biomaterials to enhance antigen-specific T cell expansion for cancer immunotherapy. Biomaterials 268, 120584, (2021) doi:10.1016 / j.biomaterials.2020.120584. 6) Sckisel, G. D. et al. Out-of-Sequence Signal 3 Paralyzes Primary CD4(+) T-Cell-Dependent Immunity. Immunity 43, 240-250, (2015) doi:10.1016 / j.immuni.2015.06.023. 7) Lambert, L. H. et al. Improving T Cell Expansion with a Soft Touch. Nano Lett 17, 821-826, (2017) doi:10.1021 / acs.nanolett.6b04071. 8) Harrison, D. L., Fang, Y. & Huang, J. T-Cell Mechanobiology: Force Sensation, Potentiation, and Translation. Frontiers in Physics 7, (2019) doi:10.3389 / fphy.2019.00045. 9) Judokusumo, E., Tabdanov, E., Kumari, S., Dustin, M. L. & Kam, L. C. Mechanosensing in T lymphocyte activation. Biophysical Journal 102, L5-7, (2012) doi:10.1016 / j.bpj.2011.12.011. 10) O'Connor, R. S. et al. Substrate rigidity regulates human T cell activation and proliferation. J Immunol 189, 1330-1339, (2012) doi:10.4049 / jimmunol.1102757. 11) Yuan, D. J., Shi, L. & Kam, L. C. Biphasic response of T cell activation to substrate stiffness. Biomaterials 273, 120797, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2021.120797. 12) Friedman, D. et al. Natural killer cell immune synapse formation and cytotoxicity are controlled by tension of the target interface. Journal of cell science 134, (2021) doi:10.1242 / jcs.258570. 13) Jansen, L. E., Birch, N. P., Schiffman, J. D., Crosby, A. J. & Peyton, S. R. Mechanics of intact bone marrow. J Mech Behav Biomed Mater 50, 299-307, (2015) doi:10.1016 / j.jmbbm.2015.06.023. 14) Qin, Q., Wang, D., Xu, L., Lan, Y. & Tong, M. Evaluating Lymph Node Stiffness to Differentiate Bacterial Cervical Lymphadenitis and Lymph Node-First Presentation of Kawasaki Disease by Shear Wave Elastography. Journal of Ultrasound in Medicine 40, 1371-1380, (2021) doi:https: / / doi.org / 10.1002 / jum.15518. 15) Grakoui, A. et al. The Immunological Synapse: A Molecular Machine Controlling T Cell Activation. Science 285, 221-227, (1999) doi:doi:10.1126 / science.285.5425.221. 16) Davis, S. J. & van der Merwe, P. A. The kinetic-segregation model: TCR triggering and beyond. Nature Immunology 7, 803-809, (2006) doi:10.1038 / ni1369. 17) Dustin, M. L. The immunological synapse. Cancer Immunol Res 2, 1023-1033, (2014) doi:10.1158 / 2326-6066.Cir-14-0161. 18) Hernandez Bücher, J. E. et al. Bottom-up assembly of target-specific cytotoxic synthetic cells. Biomaterials 285, 121522, (2022) doi:https: / / doi.org / 10.1016 / j.biomaterials.2022.121522. 19) Jia, H. & Schwille, P. Bottom-up synthetic biology: reconstitution in space and time. Curr Opin Biotechnol 60, 179-187, (2019) doi:10.1016 / j.copbio.2019.05.008. 20) Staufer, O. et al. Bottom-up assembly of biomedical relevant fully synthetic extracellular vesicles. Sci Adv 7, eabg6666, (2021) doi:10.1126 / sciadv.abg6666. 21) Staufer, O. et al. Building a community to engineer synthetic cells and organelles from the bottom-up. eLife 10, e73556, (2021) doi:10.7554 / eLife.73556. 22) Jenkins, E. et al. Reconstitution of immune cell interactions in free-standing membranes. Journal of cell science 132, (2018) doi:10.1242 / jcs.219709. 23) Göpfrich, K. et al. One-pot assembly of complex giant unilamellar vesicle-based synthetic cells. ACS Synthetic Biology, (2019) doi:10.1021 / acssynbio.9b00034. 24) Céspedes, P. F., Beckers, D., Dustin, M. L. & Sezgin, E. Model membrane systems to reconstitute immune cell signaling. Febs j, (2020) doi:10.1111 / febs.15488. 25) Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203. 26) Torres, A. J., Contento, R. L., Gordo, S., Wucherpfennig, K. W. & Love, J. C. Functional single-cell analysis of T-cell activation by supported lipid bilayer-tethered ligands on arrays of nanowells. Lab on a chip 13, 90-99, (2013) doi:10.1039 / C2LC40869D. 27) Lippert, A. H. et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal 120, 35-45, (2021) doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021. 28) Mescher, M. F. Surface contact requirements for activation of cytotoxic T lymphocytes. J Immunol 149, 2402-2405 (1992). 29) Campàs, O. et al. Quantifying cell-generated mechanical forces within living embryonic tissues. Nat Methods 11, 183-189, (2014) doi:10.1038 / nmeth.2761. 30) Ohmann, A. et al. A synthetic enzyme built from DNA flips 10 7 lipids per second in biological membranes. Nature communications 9, 2426, (2018) doi:10.1038 / s41467-018-04821-5. 31) Schuster, H. et al. The immunopeptidomic landscape of ovarian carcinomas. Proceedings of the National Academy of Sciences 114, E9942-E9951, (2017) doi:doi:10.1073 / pnas.1707658114. 32) Hammink, R. et al. Semiflexible Immunobrushes Induce Enhanced T Cell Activation and Expansion. ACS Applied Materials & Interfaces 13, 16007-16018, (2021) doi:10.1021 / acsami.0c21994. 33) Hocheng, H., Chen, C.-M., Chou, Y.-C. & Lin, C.-H. Study of novel electrical routing and integrated packaging on bio-compatible flexible substrates. Microsystem Technologies 16, 423-430, (2010) doi:10.1007 / s00542-009-0930-2. 34) Izdihar, K. et al. Structural, Mechanical, and Dielectric Properties of Polydimethylsiloxane and Silicone Elastomer for the Fabrication of Clinical-Grade Kidney Phantom. Applied Sciences 11, 1172 (2021). 35) Knörck, A. et al. Quantity, quality, and functionality of peripheral blood cells derived from residual blood of different apheresis kits. Transfusion 58, 1516-1526, (2018) doi:https: / / doi.org / 10.1111 / trf.14616. 36) Sudarsanam, H., Buhmann, R. & Henschler, R. Influence of Culture Conditions on Ex Vivo Expansion of T Lymphocytes and Their Function for Therapy: Current Insights and Open Questions. Front Bioeng Biotechnol 10, 886637 (2022). https: / / doi.org:10.3389 / fbioe.2022.886637 37) Schwaiger, S., Wolf, A. M., Robatscher, P., Jenewein, B. & Grubeck-Loebenstein, B. IL-4-producing CD8+ T cells with a CD62L++(bright) phenotype accumulate in a subgroup of older adults and are associated with the maintenance of intact humoral immunity in old age. J Immunol 170, 613-619 (2003). https: / / doi.org:10.4049 / jimmunol.170.1.613 38) Noble, A., Giorgini, A. & Leggat, J. A. Cytokine-induced IL-10-secreting CD8 T cells represent a phenotypically distinct suppressor T-cell lineage. Blood 107, 4475-4483 (2006). https: / / doi.org:10.1182 / blood-2005-10-3994 39) Mishra, S., Srinivasan, S., Ma, C. & Zhang, N. CD8+ Regulatory T Cell - A Mystery to Be Revealed. Frontiers in Immunology 12 (2021). https: / / doi.org:10.3389 / fimmu.2021.708874 40) Staufer, O. et al. Solution structure and synaptic analyses reveal molecular mechanisms of bispecific T cell engagers. bio-Rxiv, 2022.2006.2015.496334 (2023). https: / / doi.org:10.1101 / 2022.06.15.496334 41) Xu, W. et al. Cell Stiffness Is a Biomarker of the Metastatic Potential of Ovarian Cancer Cells. PloS one 7, e46609 (2012). https: / / doi.org:10.1371 / journal.pone.0046609 42) Han, Y. L. et al. Cell swelling, softening and invasion in a three-dimensional breast cancer model. Nat Phys 16, 101-108 (2020). https: / / doi.org:10.1038 / s41567-019-0680-8. 43) Hsu, C.-J. et al. Ligand Mobility Modulates Immunological Synapse Formation and T Cell Activation. PloS one 7, e32398, (2012). doi:10.1371 / journal.pone.0032398. 44) Flippe, L., Bézie, S., Anegon, I. & Guillonneau, C. Future prospects for CD8(+) regulatory T cells in immune tolerance. Immunol Rev 292, 209-224 (2019). https: / / doi.org:10.1111 / imr.12812 45) Park, S., Song, S. & Yoon, S.-H. Ultrasonication-induced and diluent-assisted suspension polymerization for size-controllable synthesis of polydimethylsiloxane droplets. Colloids and Surfaces A: Physicochemical and Engineering Aspects 644, 128827, (2022). doi:https: / / doi.org / 10.1016 / j.colsurfa.2022.128827. 46) Pan, C. et al. Silver-Coated Poly(dimethylsiloxane) Beads for Soft, Stretchable, and Thermally Stable Conductive Elastomer Composites. ACS Applied Materials & Interfaces 11, 42561-42570, (2019). doi:10.1021 / acsami.9b13266. 47) Jiang, K., Thomas, P. C., Forry, S. P., DeVoe, D. L. & Raghavan, S. R. Microfluidic synthesis of monodisperse PDMS microbeads as discrete oxygen sensors. Soft Matter 8, 923-926, (2012). doi:10.1039 / C2SM06685H. 48) Capera-Aragones, J. et al. Understanding the spatial and functional link between the IL-2R and TCR at the immunological synapse. Biophysical Journal 122, 510a (2023). https: / / doi.org:https: / / doi.org / 10.1016 / j.bpj.2022.11.2717 49) Xu, Z., Shen, C., Hou, Y., Gao, H. & Sun, S. Oleylamine as Both Reducing Agent and Stabilizer in a Facile Synthesis of Magnetite Nanoparticles. Chemistry of Materials 21, 1778-1780 (2009). https: / / doi.org:10.1021 / cm802978z QUOTES INCLUDED IN THE DESCRIPTION

[0000] This list of documents cited by the applicant was automatically generated and is included solely for the reader's convenience. The list is not part of the German patent or utility model application. The DPMA accepts no liability for any errors or omissions. Cited non-patent literature

[0000] Waldman, A. D., Fritz, J. M. & Lenardo, M. J. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nature Reviews Immunology 20, 651-668, (2020) doi:10.1038 / s41577-020-0306-5

[0001] Tantalo, D. G. M. et al. Understanding T cell phenotype for the design of effective chimeric antigen receptor T cell therapies. Journal for ImmunoTherapy of Cancer 9, e002555, (2021) doi:10.1136 / jitc-2021-002555 [0002, 0122, 0124, 0125] Zhang, D. K. Y. et al. Enhancing CAR-T cell functionality in a patient-specific manner. Nature communications 14, 506, (2023) doi:10.1038 / s41467-023-36126-7 [0002, 0125] Li, Y. & Kurlander, R. J. Comparison of anti-CD3 and anti-CD28-coated beads with soluble anti-CD3 for expanding human T cells: Differing impact on CD8 T cell phenotype and responsiveness to restimulation. Journal of Translational Medicine 8, 104, (2010) doi:10.1186 / 1479-5876-8-104 [0002, 0125] Isser, A., Livingston, N. K. & Schneck, J. P. Biomaterials to enhance antigen-specific T cell expansion for cancer immunotherapy. Biomaterials 268, 120584, (2021) doi:10.1016 / j.biomaterials.2020.120584 [0002, 0003, 0125] Sckisel, G. D. et al. Out-of-Sequence Signal 3 Paralyzes Primary CD4(+) T-Cell-Dependent Immunity. Immunity 43, 240-250, (2015) doi:10.1016 / j.immuni.2015.06.023 [0003, 0125] Lambert, L. H. et al. Improving T Cell Expansion with a Soft Touch. Nano Lett 17, 821-826, (2017) doi:10.1021 / acs.nanolett.6b04071 [0003, 0103, 0125] Harrison, D. L., Fang, Y. & Huang, J. T-Cell Mechanobiology: Force Sensation, Potentiation, and Translation. Frontiers in Physics 7, (2019) doi:10.3389 / fphy.2019.00045 [0003, 0098, 0125] Judokusumo, E., Tabdanov, E., Kumari, S., Dustin, M. L. & Kam, L. C. Mechanosensing in T lymphocyte activation. Biophysical Journal 102, L5-7, (2012) doi:10.1016 / j.bpj.2011.12.011 [0003, 0125] O'Connor, R. S. et al. Substrate rigidity regulates human T cell activation and proliferation. J Immunol 189, 1330-1339, (2012) doi:10.4049 / jimmunol.1102757 [0003, 0004, 0125] Yuan, D. J., Shi, L. & Kam, L. C. Biphasic response of T cell activation to substrate stiffness. Biomaterials 273, 120797, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2021.120797

[0003] Friedman, D. et al. Natural killer cell immune synapse formation and cytotoxicity are controlled by tension of the target interface. Journal of cell science 134, (2021) doi:10.1242 / jcs.258570 [0003, 0125] Jansen, L. E., Birch, N. P., Schiffman, J. D., Crosby, A. J. & Peyton, S. R. Mechanics of intact bone marrow. J Mech Behav Biomed Mater 50, 299-307, (2015) doi:10.1016 / j.jmbbm.2015.06.023 [0003, 0125] Qin, Q., Wang, D., Xu, L., Lan, Y. & Tong, M. Evaluating Lymph Node Stiffness to Differentiate Bacterial Cervical Lymphadenitis and Lymph Node-First Presentation of Kawasaki Disease by Shear Wave Elastography. Journal of Ultrasound in Medicine 40, 1371-1380, (2021) doi:https: / / doi.org / 10.1002 / jum.15518 [0003, 0125] Grakoui, A. et al. The Immunological Synapse: A Molecular Machine Controlling T Cell Activation. Science 285, 221-227, (1999) doi:doi:10.1126 / science.285.5425.221 [0003, 0125] Davis, S. J. & van der Merwe, P. A. The kinetic-segregation model: TCR triggering and beyond. Nature Immunology 7, 803-809, (2006) doi:10.1038 / ni1369 [0003, 0096, 0125] Dustin, M. L. The immunological synapse. Cancer Immunol Res 2, 1023-1033, (2014) doi:10.1158 / 2326-6066.Cir-14-0161 [0003, 0101, 0125] Hernandez Bücher, J. E. et al. Bottom-up assembly of target-specific cytotoxic synthetic cells. Biomaterials 285, 121522, (2022) doi:https: / / doi.org / 10.1016 / j.biomaterials.2022.121522 [0004, 0125] Jia, H. & Schwille, P. Bottom-up synthetic biology: reconstitution in space and time. Curr Opin Biotechnol 60, 179-187, (2019) doi:10.1016 / j.cop-bio.2019.05.008

[0004] Staufer, O. et al. Bottom-up assembly of biomedical relevant fully synthetic extracellular vesicles. Sci Adv 7, eabg6666, (2021) doi:10.1126 / sciadv.abg6666 [0004, 0125] Staufer, O. et al. Building a community to engineer synthetic cells and organelles from the bottom-up. eLife 10, e73556, (2021) doi:10.7554 / eLife.73556 [0004, 0125] Jenkins, E. et al. Reconstitution of immune cell interactions in freestanding membranes. Journal of cell science 132, (2018) doi:10.1242 / jcs.219709

[0004] Göpfrich, K. et al. One-pot assembly of complex giant unilamellar vesicle-based synthetic cells. ACS Synthetic Biology, (2019) doi:10.1021 / acssynbio.9b00034 [0004, 0125] Céspedes, P. F., Beckers, D., Dustin, M. L. & Sezgin, E. Model membrane systems to reconstitute immune cell signaling. Febs j, (2020) doi:10.1111 / febs.15488 [0004, 0125] Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203 [0004, 0103, 0125] Torres, A. J., Contento, R. L., Gordo, S., Wucherpfennig, K. W. & Love, J. C. Functional single-cell analysis of T-cell activation by supported lipid bilayer-tethered ligands on arrays of nanowells. Lab on a chip 13, 90-99, (2013) doi:10.1039 / C2LC40869D [0004, 0125] Lippert, A. H. et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal 120, 35-45, (2021) doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021 [0004, 0125] Mescher, M. F. Surface contact requirements for activation of cytotoxic T lymphocytes. J Immunol 149, 2402-2405 (1992

[0083] Campàs, O. et al. Quantifying cell-generated mechanical forces within living embryonic tissues. Nat Methods 11, 183-189, doi:10.1038 / nmeth.2761 (2014

[0084] Ohmann, A. et al. A synthetic enzyme built from DNA flips 107 lipids per second in biological membranes. Nature communications 9, 2426, doi:10.1038 / s41467-018-04821-5 (2018

[0084] Lippert, A. H. et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal 120, 35-45, doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021 (2021

[0084] Schuster, H. et al. The immunopeptidomic landscape of ovarian carcinomas. Proceedings of the National Academy of Sciences 114, E9942-E9951, doi:doi:10.1073 / pnas.1707658114 (2017

[0085] Hammink, R. et al. Semiflexible Immunobrushes Induce Enhanced T Cell Activation and Expansion. ACS Applied Materials & Interfaces 13, 16007-16018, doi:10.1021 / acsami.0c21994 (2021

[0085] Hocheng, H., Chen, C.-M., Chou, Y.-C. & Lin, C.-H. Study of novel electrical routing and integrated packaging on bio-compatible flexible substrates. Microsystem Technologies 16, 423-430, doi:10.1007 / s00542-009-0930-2 (2010

[0086] Izdihar, K. et al. Structural, Mechanical, and Dielectric Properties of Polydimethylsiloxane and Silicone Elastomer for the Fabrication of Clinical-Grade Kidney Phantom. Applied Sciences 11, 1172 (2021

[0086] Knörck, A. et al. Quantity, quality, and functionality of peripheral blood cells derived from residual blood of different apheresis kits. Transfusion 58, 1516-1526, doi:https: / / doi.org / 10.1111 / trf.14616 (2018

[0090] Sudarsanam, H., Buhmann, R. & Henschler, R. Influence of Culture Conditions on Ex Vivo Expansion of T Lymphocytes and Their Function for Therapy: Current Insights and Open Questions. Front Bioeng Biotechnol 10, 886637 (2022). https: / / doi.org:10.3389 / fbioe.2022.886637 [0092, 0125] Schwaiger, S., Wolf, A. M., Robatscher, P., Jenewein, B. & Grubeck-Loebenstein, B. IL-4-producing CD8+ T cells with a CD62L++(bright) phenotype accumulate in a subgroup of older adults and are associated with the maintenance of intact humoral immunity in old age. J Immunol 170, 613-619 (2003). https: / / doi.org:10.4049 / jimmunol.170.1.613 [0092, 0125] Noble, A., Giorgini, A. & Leggat, J. A. Cytokine-induced IL-10-secreting CD8 T cells represent a phenotypically distinct suppressor T-cell lineage. Blood 107, 4475-4483 (2006). https: / / doi.org:10.1182 / blood-2005-10-3994 [0092, 0125] Mishra, S., Srinivasan, S., Ma, C. & Zhang, N. CD8+ Regulatory T Cell - A Mystery to Be Revealed. Frontiers in Immunology 12 (2021). https: / / doi.org:10.3389 / fimmu.2021.708874 [0092, 0125] Lippert, A. H. et al. Soft Polydimethylsiloxane-Supported Lipid Bilayers for Studying T Cell Interactions. Biophysical Journal (2021) 120, 35-45, doi:https: / / doi.org / 10.1016 / j.bpj.2020.11.021

[0094] Staufer, O. et al. Solution structure and synaptic analyses reveal molecular mechanisms of bispecific T cell engagers. bioRxiv, 2022.2006.2015.496334 (2023). https: / / doi.org:10.1101 / 2022.06.15.496334

[0096] Xu, W. et al. Cell Stiffness Is a Biomarker of the Metastatic Potential of Ovarian Cancer Cells. PloS one 7, e46609 (2012). https: / / doi.org:10.1371 / journal.pone.0046609 [0096, 0125] Han, Y. L. et al. Cell swelling, softening and invasion in a three-dimensional breast cancer model. Nat Phys 16, 101-108 (2020). https: / / doi.org:10.1038 / s41567-019-0680-8 [0096, 0125] Hsu, C.-J. et al. Ligand Mobility Modulates Immunological Synapse Formation and T Cell Activation. PloS one 7, e32398, (2012). doi:10.1371 / journal.pone.0032398 [0098, 0125] Flippe, L., Bézie, S., Anegon, I. & Guillonneau, C. Future prospects for CD8(+) regulatory T cells in immune tolerance. Immunol Rev 292, 209-224 (2019). https: / / doi.org:10.1111 / imr.12812 [0099, 0125] Park, S., Song, S. & Yoon, S.-H. Ultrasonication-induced and diluent-assisted suspension polymerization for size-controllable synthesis of polydimethylsiloxane droplets. Colloids and Surfaces A: Physicochemical and Engineering Aspects 644, 128827, (2022). doi:https: / / doi.org / 10.1016 / j.colsurfa.2022.128827 [0103, 0125] Pan, C. et al. Silver-Coated Poly(dimethylsiloxane) Beads for Soft, Stretchable, and Thermally Stable Conductive Elastomer Composites. ACS Applied Materials & Interfaces 11, 42561-42570, (2019). doi:10.1021 / acsami.9b13266 [0103, 0125] Jiang, K., Thomas, P. C., Forry, S. P., DeVoe, D. L. & Raghavan, S. R. Microfluidic synthesis of monodisperse PDMS microbeads as discrete oxygen sensors. Soft Matter 8, 923-926, (2012). doi:10.1039 / C2SM06685H [0103, 0125] Capera-Aragones, J. et al. Understanding the spatial and functional link between the IL-2R and TCR at the immunological synapse. Biophysical Journal 122, 510a (2023). https: / / doi.org:https: / / doi.org / 10.1016 / j.bpj.2022.11.2717 [0103, 0125] Staufer, O. et al. Microfluidic production and characterization of biofunctionalized giant unilamellar vesicles for targeted intracellular cargo delivery. Biomaterials 264, 120203, doi:https: / / doi.org / 10.1016 / j.biomaterials.2020.120203 (2021

[0104] Ohmann, A. et al. A synthetic enzyme built from DNA flips 10 7 lipids per second in biological membranes. Nature communications 9, 2426, (2018) doi:10.1038 / s41467-018-04821-5 [0110, 0125] Xu, Z., Shen, C., Hou, Y., Gao, H. & Sun, S. Oleylamine as Both Reducing Agent and Stabilizer in a Facile Synthesis of Magnetite Nanoparticles. Chemistry of Materials 21, 1778-1780 (2009

[0117] https: / / doi.org:10.1021 / cm802978z

[0117] Waldman, A. D., Fritz, J. M. & Lenardo, M. J. A guide to cancer immunotherapy: from T cell basic science to clinical practice. Nature Reviews Immunology 20, 651-668, (2020) doi:10.1038 / s41577-020-0306-5.

[0125] Yuan, D. J., Shi, L. & Kam, L. C. Biphasic response of T cell activation to substrate stiffness. Biomaterials 273, 120797, (2021) doi:https: / / doi.org / 10.1016 / j.biomaterials.2021.120797.

[0125] Jia, H. & Schwille, P. Bottom-up synthetic biology: reconstitution in space and time. Curr Opin Biotechnol 60, 179-187, (2019) doi:10.1016 / j.copbio.2019.05.008

[0125] Jenkins, E. et al. Reconstitution of immune cell interactions in free-standing membranes. Journal of cell science 132, (2018) doi:10.1242 / jcs.219709

[0125] Mescher, M. F. Surface contact requirements for activation of cytotoxic T lymphocytes. J Immunol 149, 2402-2405 (1992)

[0125] Campàs, O. et al. Quantifying cell-generated mechanical forces within living embryonic tissues. Nat Methods 11, 183-189, (2014) doi:10.1038 / nmeth.2761

[0125] Schuster, H. et al. The immunopeptidomic landscape of ovarian carcinomas. Proceedings of the National Academy of Sciences 114, E9942-E9951, (2017) doi:doi:10.1073 / pnas.1707658114

[0125] Hammink, R. et al. Semiflexible Immunobrushes Induce Enhanced T Cell Activation and Expansion. ACS Applied Materials & Interfaces 13, 16007-16018, (2021) doi:10.1021 / acsami.0c21994

[0125] Hocheng, H., Chen, C.-M., Chou, Y.-C. & Lin, C.-H. Study of novel electrical routing and integrated packaging on bio-compatible flexible substrates. Microsystem Technologies 16, 423-430, (2010) doi:10.1007 / s00542-009-0930-2

[0125] Izdihar, K. et al. Structural, Mechanical, and Dielectric Properties of Polydimethylsiloxane and Silicone Elastomer for the Fabrication of Clinical-Grade Kidney Phantom. Applied Sciences 11, 1172 (2021)

[0125] Knörck, A. et al. Quantity, quality, and functionality of peripheral blood cells derived from residual blood of different apheresis kits. Transfusion 58, 1516-1526, (2018) doi:https: / / doi.org / 10.1111 / trf.14616

[0125] Staufer, O. et al. Solution structure and synaptic analyses reveal molecular mechanisms of bispecific T cell engagers. bio-Rxiv, 2022.2006.2015.496334 (2023). https: / / doi.org:10.1101 / 2022.06.15.496334

[0125] Xu, Z., Shen, C., Hou, Y., Gao, H. & Sun, S. Oleylamine as Both Reducing Agent and Stabilizer in a Facile Synthesis of Magnetite Nanoparticles. Chemistry of Materials 21, 1778-1780 (2009). https: / / doi.org:10.1021 / cm802978z

[0125]

Claims

[1] Composite particles, comprising a) a core comprising a hydrophobic phase, wherein the core is charged at the surface; b) at least one lipid bilayer enclosing the core, and the lipid bilayer comprising at least one lipid with a functional group. [2] Composite particles according to claim 1, characterized by that at least one ionic surfactant is embedded in the core. [3] Composite particles according to one of claims 1 or 2, characterized by that the lipid bilayer is a unilamellar lipid bilayer. [4] Composite particles according to any one of claims 1 to 3, characterized by that the lipid bilayer comprises phospholipids. [5] Composite particles according to any one of claims 1 to 4, characterized by that the lipid bilayer comprises phosphatidylcholine as a neutral amphiphile and at least one anionic lipid. [6] Composite particles according to any one of claims 1 to 5, characterized by, that the density of functional groups on the surface of the composite particles is between 1 and 1000 functional groups / µm 2 lies. [7] Composite particles according to any one of claims 1 to 6, characterized by , that the elastic modulus of the composite particles lies between 0.5 kPa and 3 MPa. [8] Composite particles according to any one of claims 1 to 7, characterized by that the composite particles are modified with a group for the activation of T cells. [9] Composite particles according to claim 8, characterized by that the modification is a group with affinity for CD3-CD38, or CD28 receptors. [10] Composite particles according to any one of claims 1 to 7, characterized by that the functional group is a coupling group. [11] Composite particles according to any one of claims 1 to 7, characterized by that the composite particles are modified with at least one tumor-associated antigen. [12] Use of the composite particles according to one of claims 8 and 9 for the activation of T cells. [13] Use of the composite particles according to claim 11 as cancer cell mimetics. [14] Method for producing composite particles according to any one of claims 1 to 11, comprising the following steps: a) Providing a composition comprising at least one hydrophobic phase and at least one ionic surfactant; b) Producing an aqueous dispersion of the composition by generating oil droplets; c) optional addition of at least one divalent cation to stabilize the oil droplets; d) Addition of SUVs forming a lipid bilayer comprising at least one lipid with a functional group on the surface of the oil droplets.

Images