Engineering the lymph node stroma to probe impact of interstitial fluid flow on T cell migration

Introduction: Lymph nodes (LNs) are highly organized immune organs found throughout the body that provide immunosurveillance and initiate the adaptive immune response. Within the T cell zone of the LN, a network of fibroblastic reticular cells (FRCs) maintains extracellular matrix conduits to direct incoming lymph for screening, guides T cell migration, and regulates T cell activation. Activated T cells can then egress the LN via efferent lymphatics. Interstitial flow (IF) is essential for lymph node function and is substantially increased during inflammation or immune response. However, measuring and modulating IF in vivo has remained a challenge in the LN. We sought to develop a spatially organized model of the T cell zone, incorporating FRC networks, efferent lymphatics, and pressure head-driven fluid flow, to probe how interstitial flow in the LN alters stroma and T cell activation and migration.

Materials and

Methods: Tissue engineered model: Lymph node lymphatic endothelial cells (LECs) are seeded on the underside of a tissue culture insert. FRCs are encapsulated in 0.4% methacrylated hyaluronic acid/0.2% collagen hydrogel above the LECs. Interstitial flow is created via pressure heads of cell culture media to approximate 0, 0.8, and 3 µm/s flow and maintained for up to 5 days. LN models are fixed and stained for F-actin and CD31. Quantification is performed in FIJI.

Magnetic Resonance Imaging: Models with and without stroma are prepared in tissue culture inserts and secured in a 50 mL conical tube cap. Pre-contrast T1 images are taken as a baseline. Then, 1:100 Gadolinium-DTPA (BioPal) is injected above the hydrogel at 0.5 ml/min flow rate. After 30 seconds, four post-contrast T1 images are taken. Lymph4D in MATLAB is used to produce flow maps.

Immune cells: T cells are isolated from leukocyte reduction system cones using a density gradient separation followed by immunomagnetic negative selection for target populations (STEMCELL Technologies). T cells are maintained in Immunocult T Cell Expansion Medium supplemented with interleukin-7 or with Immunocult Human CD3/CD28 T cell activator (STEMCELL Technologies). For migration studies, T cells are labeled with CFSE dye and added to empty hydrogel or LN stroma model in AIM-V universal media under 0, 0.8, and 3 µm/s flow conditions for 24 hours. Flow-through media is collected and analyzed via flow cytometry. Models are fixed and stained for F-actin and CD31.

Results, Conclusions, and Discussions: In the LN T cell zone, a network of FRCs and lymphatic vessels work together to modulate T cell migration and egress. This structure is shown in a human LN, with podoplanin+ FRCs in green and podoplanin+/CD31+ lymphatics in green and red (Fig 1A). The LN stroma model (Fig 1B) mimics this spatial separation, with LECs on the underside and FRCs in a HA/collagen hydrogel above (Fig 1C).

Representative images of FRC networks and LEC monolayers in the model are shown (Fig 2A). Under flow, FRCs form more extensive networks (Fig 2B). FRCs demonstrated significantly more proliferation under the highest flow rate (3.0 µm/s), but not 0 or 0.8 µm/s (data not shown). LECs are shown to have slight junction gaps, aligning with the function in vivo of loosening junctions to collect lymph (Fig 2C).

Naïve and activated CD4 T cells are visualized within the FRC networks (Fig 3A). Preliminary data with immune cells demonstrates that B cells, CD8 T cells, and CD4 T cells are able to egress from the model (Fig 3B). Interestingly, egress is higher from LN stroma compared to empty HA/collagen (data not shown).

Notably, presence of LN stroma in the gel was shown to alter flow paths. Using dynamic contrast enhanced imaging and pixel-based analysis, magnitude and direction of flow throughout the model is obtained (Fig 4A). This data can then be used to create streamlines, demonstrating paths of fluid movement (Fig 4B). Though velocity and diffusivity are unchanged by the presence of stroma at 24 hours (Fig 4C,D), the divergence is significantly decreased (Fig 4E). In response to IFF, FRCs also upregulate matrix-related genes, such as FN1, COL1A1, VCAN, and COL4A2 (data not shown). The remodeling of the matrix and flow paths by FRCs could contribute to altered migration of T cells.

These results demonstrate a novel model of the T cell zone to study the response of LN stroma and immune cell to interstitial flow. Next steps include analyzing T cell proliferation and IFNgamma secretion under flow and in the LN stroma and examining the impact of T cells on the FRC network.

Acknowledgements (Optional):