Marking true T_FH cells

Hai Qi describes a 2007 paper by Jason Cyster and colleagues that identified PD1 as a marker for T follicular helper cells.

For students in my lab who want to study T follicular helper (T_FH) cells, a paper by Jason Cyster and colleagues published in 2007 is a must-read. To help B cells to form germinal centres (GCs), T cells need to relocate into the B cell follicle and then into the GC. Although the follicular homing chemokine receptor CXCR5 had long been implicated in this process, it had not been fully characterized how exactly T cells accomplish those relocation steps. Another important question at the time was whether GC-associated T cells are different from T cells found elsewhere in the body and whether any marker could uniquely identify such cells by flow cytometry, the favorite tool of immunologists owing to its much higher processing throughput than histology or live imaging.

The study by Haynes et al. provides detailed characterization of how a combination of CXCR5 upregulation and CCR7 downregulation is important for T cells to be efficiently positioned in the follicle, and it also identifies PD1 as a marker for the CXCR5^hiCCR7^low population of GC-associated T cells. This link between tissue location and surface phenotype greatly facilitated the further characterization of T_FH cells, and PD1 has since become an essential part of the surface-phenotyping panel for T_FH cells.

However, in more recent literature, the use of PD1 as a marker for T_FH or T_FH-like cells among activated CD4⁺ T cells in many other settings is so common that it can cause confusion. Haynes et al. were very careful to describe the context of their findings as GCs induced by immunization and GCs in the Peyer’s patch. In that context, assigning PD1 as a marker for GC-associated T cells was backed up by immunohistology. In other settings, whether and how PD1 expression relates to T cell positioning or particular helper functions are questions that have to be answered on an individual basis. This is a crucial point that many students initially do not realize.

Another interesting set of observations by Haynes et al. was that T cell-specific CXCR5 ablation almost completely prevented T cells from correct positioning in the follicle, but did not prevent T cells from accessing GCs and only caused a twofold reduction in the GC response, a finding that was also reported in a separate study by Arnold et al. It remains an open question how CXCR5-deficient T cells enter GCs. What do these observations tell us about the essence of T_FH cells, if they are defined only as CXCR5^hiPD1^hi cells? That is also a question that students of today must think hard about.