Ligand Nano-cluster Arrays in a Supported Lipid Bilayer

Ligand Nanocluster Array Enables Artificial-Intelligence-Based Detection of Hidden Features in T-Cell Architecture

Protein patterning has emerged as a powerful means to interrogate adhering cells. However, the tools to apply a sub-micrometer periodic stimulus and the analysis of the response are still being standardized. We propose a technique combining electron beam lithography and surface functionalization to fabricate nanopatterns compatible with advanced imaging. The repetitive pattern enables a deep-learning algorithm to reveal that T cells organize their membrane and actin network differently depending upon whether the ligands are clustered or homogeneously distributed, an effect invisible to the unassisted human eye even after extensive image analysis. This fabrication and analysis toolbox should be useful, both together and separately, for exploring general correlation between a spatially structured subcellular stimulation and a subtle cellular response.

T Cells on Engineered Substrates: The Impact of TCR Clustering Is Enhanced by LFA-1 Engagement

We created APC-mimetic synthetic substrates to study the impact of ligand clustering on T cell activation and spreading. The substrates exhibit antibodies directed against the TCR-complex in the form of a patterned array of sub micrometric dots surrounded by a fluid supported lipid bilayer (SLB) which may itself be functionalized with another bio-molecule. We show that for T cell adhesion mediated by T cell receptor (TCR) alone, in the patterned, but not in the corresponding homogeneous controls, the TCR, ZAP-70 and actin are present in the form of clusters or patches that co-localize with the ligand-dots. However, global cell scale parameters like cell area and actin distribution are only weakly impacted by ligand clustering. In presence of ICAM-1 - the ligand of the T cell integrin LFA-1 - on the SLB, the TCR is still clustered due to the patterning of its ligands, but now global parameters are also impacted. The actin organization changes to a peripheral ring, resembling the classical actin distribution seen on homogeneous substrates, the patterned membrane topography disappears and the membrane is flat, whereas the cell area increases significantly. These observations taken together point to a possible pivotal role for LFA-1 in amplifying the effect of TCR-clustering. No such effect is evident for co-engagement of CD28, affected via its ligand B7.2. Unlike on ICAM-1, on B7.2 cell spreading and actin organization are similar for homogeneous and patterned substrates. However, TCR and ZAP-70 clusters are still formed in the patterned case. These results indicate complementary role for LFA-1 and CD28 in the regulation and putative coupling of TCR micro-clusters to actin. The engineered substrates presented here clearly have the potential to act as platform for fundamental research in immune cell biology, as well as translational analyses in immunotherapy, for example to screen molecules for their role in T cell adhesion/activation.