Cutting Edge: Synapse Propensity of Human Memory CD8 T Cells Confers Competitive Advantage over Naive Counterparts

Abstract

Memory T cells are endowed with multiple functional features that enable them to be more protective than naive T cells against infectious threats. It is not known if memory cells have a higher synapse propensity (SP; i.e., increased probability to form immature immunological synapses that then provide an entry into different modes of durable interaction with APCs). In this study, we show that only human memory CD8 T cells have remarkably high SP compared with naive counterparts. Such a dichotomy between naive and memory cells is not observed within the human CD4 or murine CD8 T cell population. Higher SP in human memory CD8 T cells allows them to outcompete and prevent naive CD8 T cells from getting recruited to the response. This observation has implications for original antigenic sin and aging of the immune system in humans.

FIGURE 1.

CD8⁺ hTm cells have high SP on uniform stimulatory surfaces. (A) CD8⁺ hTm (in green) and naive (in red) cells forming IS or IK (dark patches) on bilayers presenting 0.3 UCHT1 molecules/μm². Attachment, along with additional criteria, is used to count cells with IS or IK. Such cells are highlighted with a white triangle in this example. (B) Centralized accumulation of Alexa 568-UCHT1 Fab′ at 30, 3, and 0.3 molecules/μm² on bilayers in fixed CD8⁺ hTm cells visualized by TIRF. (C and D) Fractions of naive and memory cells in the human CD8 (C) and CD4 (D) subsets forming IS or IK on bilayers at indicated densities of UCHT1 Fab′, with expected number of cells per field (typically 30) as the denominator. Eight to ten fields per condition and subset were imaged and plotted. The data shown are representative of two independent experiments. (E–H) Fractions of human naive and memory T cells forming IS or IK on uniformly coated surface with immobilized CCL21, ICAM1, and OKT3. Note the differences in OKT3 concentration [(E) versus (F)] used for coating, presence of anti-CD28 and anti-CD8 Abs (G), and the data for the CD8 versus CD4 subsets (H). Time-lapse data (1.5–2 h) was used to determine the number of cells forming IS or IK based on scoring for attachment and deceleration. Each data point represents a separate donor and independent experiment. Mean value is shown in red. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.

FIGURE 2.

CD8⁺ hTm cells have high SP on stimulatory spots. (A) Schematic of the microcontact printing procedure. Ab is first adsorbed to polydimethylsiloxane (PDMS) casts that have short pillars with flat tops that will generate the spots. When the cast is laid on the coverglass, some of the Ab gets transferred from the PDMS surface onto the coverglass. However, this only happens at the top of the pillars. CCL21 and ICAM1 are then adsorbed uniformly across the surface from solution. (B) Number of CD8⁺ hTm and naive cells arresting on 10-μm spots that are 30 μm apart over 90 min. Sixty-four spots and typically at least the same number of cells were present in the field. (C and D) Rate of attachment (on-rate) of human naive and memory CD8 T cells on 10-μm spots that are 30 μm apart (C) and 20-μm spots that are 50 μm apart (D). The on-rate represents the values for the imaging field (50,625 μm²). (E and F) Rate of encounter of human naive and memory CD8 T cells with 10-μm (E) and 20-μm (F) spots. Again, the values given are for the imaging field, as above. (G and H) Arrest efficiency of human naive and memory CD8 T cells on 10-μm (G) or 20-μm (H) spots. Both encounter rate and arrest efficiency contribute to very high on-rate of attachment of CD8⁺ hTm cells compared with the naive cells. Each data point represents a separate donor and independent experiment in each plot. Mean value is shown in red. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 3.

CD8⁺ hTm cells compete out naive T cells by preventing access to stimulatory spots because of high SP. (A–D) “1×” number (see Materials and Methods) of differentially labeled naive and memory cells were pooled and introduced into the channel with stimulatory spots. After 2 h, multiple fields along the channel were imaged for multiple donors. Mean number of cells attached per spot is plotted for each field: human CD8 T cells in (A) and (B), human CD4 T cells in (C), and murine CD8 T cells in (D). Mean value is shown in red. (E) Activation of human naive CD8 T cells assessed after 10–12 h of interaction with 10-μm stimulatory spots in a competitive setting by staining for CD62L. Having “2×” number of naive cells is a control case (blue) to compare against the scenario in which memory cells are also present (orange). Note that the histograms have the same number of naive cells represented. (F–H) Percentage of activated naive CD8 T cells, as measured by the percentage of cells with lower CD62L, under different competitive settings on 10-μm spots (F), 20-μm spots (G), and uniformly stamped surface (H). Lines that join the dots for each competitive setting represent the same donor. Data points for each competitive setting are color-coded [see top of (E) and (G)]. “1×” number of cells on 20-μm spots is considerably higher (see Materials and Methods). Channels with uniformly stamped surface received the same number of cells as the channels with 10-μm spots under the corresponding competitive settings. Each data point represents a separate donor and independent experiment in each plot. *p < 0.05, **p < 0.01, ***p < 0.001, ****p < 0.0001.