Editors' ChoiceImmunotherapy

10,368 first dates: Microfluidic T cell matchmaking

See allHide authors and affiliations

Science Translational Medicine  21 Nov 2018:
Vol. 10, Issue 468, eaav9144
DOI: 10.1126/scitranslmed.aav9144


Coencapsulation of T cells and target cells into arrays of subnanoliter droplets improves the identification of functional T cell receptors.

Nature has evolved a staggering amount of diversity into T cell receptors (TCRs), enabling the adaptive immune system to recognize the plethora of targets that it may encounter. It is estimated that there are trillions of unique TCRs, and because T cells express only one TCR sequence each, the odds that a specific cell binds and activates a given T cell are very low. This presents a profound challenge for immunotherapy approaches that need to identify TCRs with appropriate immunogenic activities. A recent report by Segaliny et al. aims to improve the selection of TCRs in what amounts to a high-throughput cell-cell dating service.

The group first generated a target cell line expressing tumor antigen NY-ESO-1 and two inducible GFP reporter T cell lines that had affinity either to NY-ESO-1 or to another tumor antigen. Target cells and T cells were coencapsulated into subnanoliter droplets using a microfluidic droplet generator. At a scale that reality TV producers could only dream of, these droplets were then deposited into arrays of 10,368 wells for continuous observation of paired cells over the course of 9 hours. The team tracked the kinetics of GFP activation to identify activated NY-ESO-1 binding partners without identifying any false positives (TCRs transduced with non–NY-ESO-1 antigen). Laser-induced cavitation was then used to selectively displace and recover individual droplets from which the NY-ESO-1 TCR sequence was recovered. Thus, TCRs that activated an immunogenic response in cell partners could potentially be identified and later engineered into personalized T cell therapies.

Although this is an elegant demonstration of the utility of droplet microfluidics, some limitations of this proof-of-concept study need to be addressed. It has yet to be shown that this technology works with patient-derived target cells and T cells, which contain far more clonotypes than demonstrated with the two TCR reporter cell lines used here. Furthermore, the use of transduced GFP expression as a readout for T cell activation is slow and likely impractical when working with patient-derived cells. Nonetheless, the accurate identification of high activity TCRs by high-throughput T cell screening in automatable workflows could lead to faster and more effective patient-specific immunotherapies. It appears that finding compatible binding partners in the lab, as with love, may simply be a numbers game.

Highlighted Article

View Abstract

Navigate This Article