The struggle to make an effective HIV vaccine is ongoing. In recent years the field has taken a step back and started to examine the immune response to vaccination in hopes of understanding what response is needed for protection. Current dogma is that a strong B cell response is needed for protection. However, you cannot have a strong B cell response without helper T cells. Interactions between B and T cells take place in the germinal centers (GC), which are secondary lymphoid tissues where B cell proliferation and differentiation take place. In order for this process to occur, B cells need to interact with GC CD4+ T follicular helper cells (Tfh), which support selection and expansion of high affinity B cells. Studies have shown that a lack of Tfh cells led to impaired B cell response and GC formation. Therefore B/T cell interactions are important for vaccine success and should be studied in the context of HIV vaccination. In a paper recently published in Journal of Experimental Medicine, Dr. Antje Heit and colleagues in the Vaccine and Infectious Disease Division at Fred Hutch sought to identify Tfh cell functions both before and after vaccination in the context of tonsillar tissue and matched blood. In previous murine vaccination model studies it has been shown that GC Tfh cells can exit the GC and enter the circulating T cell pool. However, whether this happens in humans was unknown. Recently, a population of circulating CXCR5+ CD4+ memory T cells was identified and found to help B cells, yet it is still unknown if these cells originated in the GC. In this study, the researchers hoped to better understand the interplay between B and T cells to determine the role of Tfh cells in vaccine-induced immune responses. If these circulating Tfh cells play a role in this response, they may serve as a circulating marker for the GC immune response. In Dr. Heits’ words, “The aim of the study is to test the hypothesis whether vaccine-induced, circulating memory Tfh cells can serve as correlates of vaccine efficiency and predict a long lived B cell and thus antibody response. Circulating Tfh cells are still not fully understood and a definitive link to the GC Tfh is missing. Thus our work, for the first time, presents some evidence that within an individual, circulating CXCR5+CD4+ memory Tfh cells are clonally related to actual GC Tfh cells”.
Figure provided by Dr. Heit.
In order to identify Tfh cells the group used cell surface markers such as CXCR5, a B cell follicle-homing chemokine receptor, PD-1 a T cell co-inhibitory receptor and the co-stimulatory molecule ICOS. In this study CXCR5+ CD4+ T cells were split into 3 categories: 1. PD1+ ICOS+ (DP) 2. PD1+ ICOS- (PD1) and 3. PD1- ICOS- (DN). Phenotypes of each category were analyzed by cell surface staining and compared to GC Tfh cells. Of the three groups the DP group was found to be the most phenotypically similar to GC Tfh cells. Looking more closely at similarities between the groups and GC Tfh cells, eight tonsil and matched blood samples were used for T cell receptor b (TCRb) sequencing. This allows for identification of clonal similarities between groups, which would suggest similar origins. This analysis found that PD1+ cells (both DP and PD1 groups) were most clonally related to the GC Tfh cells (see figure). This similarity in both phenotype and clonality suggests a model where GC cells can leave the GC after antigen activation and enter the circulating Tfh (cTfh) cell population. In order to test this model, the group used blood collected pre- and post-vaccination from three HIV vaccine candidates, 1. Adenovirus/HIV prime boost 2. DNA prime plus adenovirus/HIV boost and 3. gp120 (HIV envelope) prime boost. They hypothesize that the vaccine- specific cTfh cell population should change after vaccination if their model is correct. Looking at the three populations the group found that with any vaccine strategy only the DP cTfh cells expanded after vaccination with the other two populations remaining steady. In addition to population frequencies the category of the T cells were identified. Most CXCR5+ CD4+ T cells are memory cells, which become activated, differentiate into effector cells, and migrate to the T/B cell border by down-regulating CCR7. To see if this is the case with the cTfh cells, the group characterized and followed these cells. They found that the DP cells were an equal mix of T central memory cells (Tcm) and T effector memory (Tem) cells with low CCR7 expression, the PD1 group was mostly CCR7+ Tcm cells and the DN group was composed of mostly Naive T cells and that distribution of these populations did not change with vaccination. They also characterized CD127 expression, which identifies Tfh cells outside the GC. Similar to CCR7, the DP population expressed the lowest frequency of cells expressing CD127 compared to the other two groups.
In order to identify if the differences seen in the DP population are a direct effect of TCR-mediated recognition by vaccine antigens, the group looked at activation of these cells by vaccine-matched peptides. Looking at activation markers, the group found an increase in CD40L (an activation marker) expression in the HIV-specific DP cells. In addition to activation markers, transcription levels of IL-21 and CXCL13 (GC Tfh cell cytokines) were measured. Again, the DP population had increases in both markers after vaccination, suggesting activation and a GC Tfh cell-like program. To more closely follow the GC-like cells, the TCR b sequences were analyzed to look for clonality and HIV-specific T cell selection after vaccination. From this they found vaccine-induced PD1+ cTfh subsets were clonally related which supports the existence of a persistent clonal relationship that is influenced by vaccination (see figure). This also suggests, that cells are moving from the PD1 Tcm pool after activation to the DP Tem pool.
To tie the various T cell effects seen after vaccination to B cells, the group looked for correlations between the HIV-specific DP population and B cell populations. They found that both populations had an increase in plasmablasts (antibody producing cells) and HIV- specific IgG. Overall this study supports a model where primed GC Tfh cells can exit the GC to establish a pool of cTfh cells that can reactivate with re-exposure having GC Tfh-like properties. The authors hypothesize that this allows for the cTfh population to be used as a biomarker for GC activity and may be helpful in assessing viable vaccine design candidates.
Heit A, Schmitz F, Gerdts S, Flach B, Moore MS, Perkins JA, Robins HS, Aderem A, Spearman P, Tomaras GD, De Rosa SC, McElrath MJ. 2017. Vaccination establishes clonal relatives of germinal center T cells in the blood of humans. J Exp Med.
This work was supported by the National Institutes of Allergy and Infectious Disease, National Institutes of Health.