Building a microphysiological model of the human lymph node follicle border zone

Introduction: The human lymph node is central to vaccine development, and is where antibody responses to vaccine antigens are initiated. Immune cell subsets such as T follicular helper (Tfh) and memory B (memB) cells, which reside in T cell zones and B follicles, respectively, cross-talk in a spatially-regulated manner to initiate germinal center development and antibody production. Specifically, naïve T cells, which reside in the T zone through CCL21 chemotaxis, are first activated through engagement of their T cell receptor. A subset of activated T cells are further differentiated into Tfhs. Tfhs and memBs briefly meet at the border of these zones at an essential turning point in the adaptive response (Fig. 1a), but the dynamics and soluble cues responsible for the spatial regulation are poorly understood. As human lymph nodes are rarely available and traditional in vitro cell culture approaches remove all spatial features, an organized microphysiological system with human immune lymphocytes (LN-MPS) is needed for progress. To model spatial arrangement of the border zone, we designed the LN-MPS to have three adjacent and abutting 3D-culture channels for a T zone, border zone, and B zone, respectively (Fig. 1d, i).


Materials and Methods: Naïve CD4+ T lymphocytes and memory B lymphocytes were isolated from autologous donors (StemCell) using immunomagnetic separation. Naïve CD4+ T cells were cultured in presence of activating anti-CD3/28 antibodies along with Activin A, IL-12, and IL-7 for 4 days to stimulate T follicular helper cell differentiation. MemB and Tfh cells were labeled with NHS rhodamine (1 µM) and Calcein AM (2 µM), respectively, prior to integration on chip. Cells were resuspended in thermally-setting collagen-fibrinogen hydrogel (Ibidi). To fabricate the three-lane microfluidic chip, SU8 masters were constructed via photolithography. PDMS layers and glass were activated with ozone exposure and then bonded to make chips. Surface tension barriers between the hexagonal posts enabled filling of the outer two gel lanes first. Following the first gelation step, the media lanes and central gel lane were filled next. After a brief rest, media lanes were exchanged with the appropriate experimental treatment (i.e. T cell activator or CCL21 in AIM5 media). To establish a gradient, chemokine was added to the left lane only. The viability assessment was performed via Calcein AM (live) and DAPI (dead) stain. Live and dead cells were counted with ImageJ. To assess T cell activation, naïve T cells were activated on chip with Immunocult CD3/CD28 ligand, and IFNy ELISA was performed on media sampled from media reservoirs with no dilution.


Results, Conclusions, and Discussions:

To demonstrate that sensitive primary human lymphocytes remain functional in the LN-MPS, we assessed T cell viability, activation, and chemotaxis. In preliminary work, we found that CD4+ T cells were particularly vulnerable to oxygen depletion in the chip, so large open reservoirs were placed on each end of the media channels. In the optimized design, the viability of naïve CD4+ T cells on-chip after four days of culture was confirmed by Calcein/DAPI staining (78±2 %, n=3 chips). To determine if naïve T cells on chip responded appropriately to an activation signal, we added CD3/CD28 tetramer (Immunocult) to the media reservoirs, and observed the expected strong IFNγ secretion after 5 days (Fig. 1b). To confirm that patterned cells could respond chemotactically in the on-chip 3D culture, we added CCL21 to the left media reservoirs to form a gradient, and observed cell migration toward the chemokine within 1 hour, matching biological timescales (Fig. 1c). Thus, this patterning strategy was sufficient to provide the micropatterned culture that is required to mimic T cell function and local gradients of signals that are found in the lymph node. We continue to characterize the system in terms of B cell responses, and interactions between T cells and B cells on the chip. As a first test, we positioned Tfh cells separately from memB cells, and observed memB cell migration into the Tfh zone (Fig. 1d).



Conclusions

We developed a three-lane microfluidic chip with partial walls that enables surface tension-based filling, with the goal of mimicking the arrangement of the B/T border zone. In the optimized chip design, naïve human T cells remained viable, responded to stimulus, and chemotactically migrated following integration on the microfluidic chip. The results show distinct lymphocyte populations were initially positioned separately, and then migrated across lanes during culture. We are currently using this system to investigate mechanisms of migratory behavior and cross-talk of memB and Tfh cells at the follicle border. This research laid the groundwork for generation of MPS systems with organized lymphocyte compartments to assess the role of spatial organization in the adaptive immune response.


Acknowledgements (Optional): This work was supported by the National Institute of Biomedical Imaging and Bioengineering (NIBIB) under Award Number U01EB029127 through the National Institutes of Health (NIH), with co-funding from the National Center for Advancing Translational Sciences (NCATS).


References (Optional): :