Fig 1.
Mature effector T cells decay faster than effector memory T cells in P. chabaudi.
Decay of malaria-specific polyclonal cells was detected by infecting C57Bl/6 mice with P. chabaudi and administration of BrdU either (A-E) during the memory phase (days 24–30), or (F-K) the peak of infection (days 4–10) to label Teff. The day after the end of BrdU administration, the infection was terminated with mefloquine (MQ), and decay of CD4+ memory (CD44hi CD127hi CD11a+, BrdU+/-) and effector (CD127-, (CD11a, BrdU)+/-) T cells in the spleen was determined by flow cytometry. A) Schematic representation of the experimental design for memory phase. B) Plots show the gating strategy for Tmem, including BrdU. Graphs showing C) number of Tmem (CD4+ CD11a+ CD44hi CD127hi) and D) survival of memory T cell subsets in the spleen after parasite clearance. E) Graphs showing percentage (left) and number (right) of cells in each Tmem subset that survive after proliferating days 24–30 (BrdU+). F) Schematic representation of the experimental design to study decay of effector T cells, where BrdU was given days 4–10 p.i., and infection was terminated with MQ days 10–14 p.i. Flow cytometric gating and graphs showing G) number of Teff (CD4+ CD127-) and H) survival of Teff subset populations in the spleen after parasite clearance. I) Plots showing the gating strategy for Teff from CD11a+ to Teff (CD4+ CD11a+ CD44hi CD127lo). Graphs showing J) the number of divided Teff (CD4+ CD11a+ CD127- BrdU+), as they decay after labelling days 4–10 p.i, and subsequent parasite clearance, and K) the percentage (left) and number (right) of cells in each of the divided Teff subsets as they decay. Data represent 3 mice per group. Data was analyzed by Student’s t test and error bars represent SEM. * represents a significant difference between subsets at one timepoint. † represents a significant difference between timepoints d10-d30, or d30-50, # from d20-30 only; one symbol p<0.05, two symbols p<0.01, n.s.–not significant with color coding of symbols to indicate which subset changes.
Fig 2.
TeffEarly survive like Tmem cells, while highly activated Teff subsets decay.
A) Schematic representation of the experimental design. T cell subsets were sorted from spleens of infected B5 TCR Tg animals (Effector on d8 p.i. (top) and Memory on d60 p.i. (bottom)) and the same number of T cells of each subset (5 x 104) were transferred into uninfected congenic recipients (Thy1.1) for 60 days. B) Graph showing numbers of B5 TCR Tg (CD4+ Thy1.2+) T cells recovered from spleens of recipients of each T cell subset on d60 post-transfer. Data represent 4–9 mice per group from two experiments. Data were analyzed using Student’s t test, **p<0.01, *p<0.05. n.s.–not significant. C) Concatenated contour plot of B5 T cells recovered from TeffEarly recipients day 60 post-transfer showing memory (CD44, CD127) and memory T cell subset (CD62L, CD27) phenotype, and summary graph of Tmem phenotype of recovered T cells from the groups of mice that received each Teff subset. Error bars represent SEM.
Fig 3.
CD4 Memory T cell subsets differentiate from Tcm to TemLate even after elimination of chronic infection.
A) Schematic representation of the experimental design. (top) Memory T cell subsets from infected B5 TCR Tg donors (Thy1.2), that were not treated (A-D), or had been treated (bottom, E, F) with chloroquine (CQ) on days 30–34, were sorted from the spleen (d60 p.i.), and transferred (2.5 x 105) into uninfected congenic (Thy1.1) recipients for 14 days. B) Concatenated contour plots with outliers showing the memory phenotype (CD44, CD127) of all B5 T cells (CD4+ Thy1.2+) recovered from all animals in each group. C) Numbers of B5 T cells recovered from spleens of recipients of Tmem from untreated donors post-transfer. D) Concatenated contour plots and summary stacked bar graph showing memory subset phenotypes (CD62L, CD27) of all B5 memory T cells recovered from all animals in each group. Summary bar graph shows average in each Tmem subset gate on recovery. (E, F) Memory T cell subsets sorted from infected mice treated with anti-malarial drug, CQ, were recovered from uninfected recipients. E) Numbers of B5 T cells recovered from recipients of B5 Tmem from CQ treated donors are shown. F) Concatenated contour plots and summary graph with outliers showing subset phenotypes of B5 Tmem recovered from recipients of B5 Tmem from CQ treated donors. Summary bar graph shows average in each Tmem subset gate on recovery. Data represent 2–3 mice per group. Error bars represent SEM, and n.s–no significant difference between all groups, ***p<0.001 comparing the distribution of the subset phenotypes of B5 T cells recovered in the three groups of recipients, #p<0.05 comparing the fraction of chronically stimulated vs. rested TemEarly donor cells recovered as TemLate in panels D and F.
Fig 4.
Persistent infection promotes Tem survival.
A) Schematic representation of the experimental design. Memory T cell subsets were sorted from infected B5 TCR Tg mice (d60 p.i.) and the same number of each subset (2.5 x 105) were transferred into either infection-matched (d60 p.i., top) or uninfected (bottom) Thy1.1 hosts for 60 days after transfer. B) Graph shows numbers of recovered B5 T cells (CD4+ Thy1.2+) in recipients of TemEarly. C) Concatenated contour plots show division (CFSE-), and levels of CD127, and phenotypes (CD62L, CD27) of cells recovered from TemEarly recipients. Summary bar graphs show average of divided cells or fraction in each Tmem subset gate on recovery. D) Graph shows numbers of recovered B5 T cells in recipients of TemLate. E) Concatenated contour plots show levels of CFSE, and CD127, and phenotypes (CD62L, CD27) of cells recovered from TemLate recipients. Summary bar graphs show average of divided cells or fraction in each Tmem subset gate on recovery. Data are representative of 2–4 mice per group from 2 similar experiments. Data was analyzed by Student’s t test, and error bar represents SEM, *p<0.05, n.s.–not significant. On summary bar graphs symbols refer to differences between fractions of recovered cells in each Tmem subset in a given stacked bar, one symbol = p<0.05; two symbols, **p<0.01 comparing Tcm to TemL; three symbols, ***p<0.001 comparing TemEarly to TemLate, # comparing Tcm (very few) to TemEarly, + comparing TemEarly to TemLate.
Fig 5.
Late effector memory T cells protect immunodeficient animals from malaria.
A) Schematic of the experimental design. Memory T cell subsets were sorted from spleens of infected B5 TCR Tg mice on d60 p.i, and transferred (2 x 105) with immune B cells (2 x 107) into RAG2o mice. Recipient mice were then infected with P. chabaudi (5x104 iRBC). Parasitemia, weight, and temperature were assessed daily for 40 days p.i., and splenocytes were analyzed by flow cytometry on day 40. Graphs of the average B) peak parasitemia, and percent change of C) weight and hypothermia of recipient mice at the peak of each symptom for each recipient (d8-10 p.i.). Flow cytometric analysis summarized here as D) total cell numbers of B5 T cells (CD4+Thy1.2+) recovered and E) the percent of B5 T cells that are IFN-γ+. Data shown represent 3 mice per group and are representative of 5 independent experiments. Error bar represents SEM, *p<0.05, **p<0.01, n.s.–not significant.
Fig 6.
All Effector T cell subsets protect from parasitemia.
A) Schematic of experimental model. Effector T cell subsets were sorted from spleens of B5 TCR Tg on d8 p.i., and transferred (5x105) with immune CD19+ BALB/c B cells (1x107) into RAG2o mice that were then infected with P. chabaudi (5x104 iRBC). Parasitemia and pathology were followed for two weeks. Graphs showing average B) peak parasitemia (%iRBC/RBC) summarized from two experiments (n = 6), C) % change of weight, and hypothermia of recipient mice at the peak of each symptom for each recipient (d8-10 p.i.). Flow cytometry of splenocytes was done on day 14 p.i. and D) graph shows average number of B5 T cells (CD4+Thy1.2+) recovered. E) Histograms, contour overlay (TeffLate), and summary graphs of cytokines produced by T cells from recipients of each Teff subset. Data show 2–3 mice per group and are representative of 3 independent experiments. Error bars show SEM, *p<0.05, **p<0.01, ***p<0.001, n.s–not significant.
Fig 7.
Rested effector T cells protect from parasitemia and pathology.
A) Schematic of the experimental design. Effector T cell subsets were sorted from the spleens of B5 TCR Tg at d8 p.i. and transferred (5 x 105) into RAG2o mice for 60 days before infection. Recipients were then infected with 105 P. chabaudi. Parasitemia, weight, and temperature change was measured for 14 days after infection. B) Average peak parasitemia (%iRBC/RBC) of each mouse (d8-10 p.i.) is shown. As measurements of pathology, the average of C) weight and D) temperature loss (% change) of each recipient at the peak of pathology (d8-10 p.i.) are shown. Flow cytometry was performed to quantitate E) the number of B5 T cells (CD4+Thy1.2+) recovered from RAG2o recipients on day 14 p.i. and F) graph showing MFI of IFN-γ in T cells recovered from each group. Data represent 2–3 mice per group and are representative of 3 similar independent experiments. Parasitemia was analyzed using one-way ANOVA and Tukey’s post-hoc. Error bar represents SEM, * p<0.05, ** p<0.01, *** p<0.001, n.s—not significant.
Fig 8.
Proposed model of T cell activation and memory T cell differentiation.
Upon activation of naïve T cells, TeffEarly (CD127- CD62Lhi CD27+) are generated, which contain precursors to both effector and memory T cells. Depending on the presence of antigen (or leukopenia) in their environment, they can progress toward maturation of effector T cells, TeffInt, TeffLate subsets, or in an uninfected environment, towards differentiation into Tcm and then Tem subsets, the latter being promoted by low-level chronicity. Upon downregulation of CD62L, Teff lose survival potential and become terminal Teff; however, surviving mature Teff can become CD127hi, and expand when re-activated and promote protection. Tmem can proliferate without downregulating CD127 in conditions of low antigen exposure (Fig 4B). Dotted arrows show less common events, or those dependent on environmental changes. Tmem can become Teff again in conditions of re-exposure to higher antigen loads (S2B Fig). We have also observed the more terminally differentiated subsets re-expressing CD62L, suggesting plasticity in this process. The degree of differentiation to central or effector memory T cells from TeffEarly is determined in the first week of P. chabaudi infection [18]. In chronic infection, and for a period after exposure to long-lasting infection, Tcm can continue to generate Tem. Populations that protect best are marked by green asterisks, and populations that survive best are marked by red asterisks.