Tissue-resident memory T cells trigger rapid exudation and local antibody accumulation

Abstract

Adaptive immunity is didactically partitioned into humoral and cell-mediated effector mechanisms, which may imply that each arm is separate and does not function together. Here, we report that the activation of CD8+ resident memory T cells (T_RM) in nonlymphoid tissues triggers vascular permeability, which facilitates rapid distribution of serum antibodies into local tissues. T_RM reactivation was associated with transcriptional upregulation of antiviral signaling pathways as well as Fc receptors and components of the complement cascade. Effects were local, but evidence is presented that T_RM in brain and reproductive mucosa are both competent to induce rapid antibody exudation. T_RM reactivation in the mouse female genital tract increased local concentrations of virus-specific neutralizing antibodies, including anti-vesicular stomatitis virus, and passively transferred anti-HIV antibodies. We showed that this response was sufficient to increase the efficacy of ex vivo vesicular stomatitis virus neutralization. These results indicate that CD8+ T_RM antigen recognition can enhance local humoral immunity.

Fig. 1

T_RM reactivation induces immune-activating transcriptional changes. (A) Schematic of experimental set-up. P14 immune chimeras 120 days post-infection were challenged transcervically with gp33 or control SIINFEKL peptide and microarray was performed 9 hours later; (B) Volcano plot of differentially expressed genes in FRT where T_RM were recalled versus control; (C) Top 100 differentially expressed genes categorized by function; (D) Fc receptor and complement gene expression in FRT tissue where T_RM were recalled versus control. Data from n = 3 mice in each group; experiment performed once; (E) Representative PIGR staining in the uterine horns of Depo-Provera treated P14 immune chimera mice 35 days post-infection transcervically challenged with gp33 or control SIINFEKL peptide. Tissue taken 6 hours after peptide challenge. PIGR in magenta, DAPI in blue. < scale bar = 500 μm>; (F) quantification of E (n = 5 mice for T_RM recall and n = 6 for control). Data are pooled from two independent experiments. Statistical significance was determined by unpaired two-tailed Mann–Whitney test where *p < 0.05. Error bars indicate standard error of mean. DAPI = 4′,6-diamidino-2-phenylindole dihydrochloride; LCMV = lymphocytic choriomeningitis virus; PIGR = polymeric immunoglobulin receptor; T_RM = Tissue-resident memory T cells; WT = wild-type.

Fig. 2

T_RM reactivation induces changes in IV antibody staining in non-lymphoid tissue. (A) P14 immune chimeras were challenged transcervically with gp33 peptide (n = 6 mice) or control peptide (SIINFEKL; n = 6 mice) then 6 hours later biotinylated anti-CD8α antibody was transferred intravenously and mice were euthanized after 3 minutes. Representative flow cytometry plots from FRT tissue. Cells gated on CD8β +. Quantification of mean fluorescent intensity of IV-negative stained cells (gate indicated in histogram); (B) Levels of IgG2c, IgA, and IgM in the vaginal lumen of P14 immune chimera mice, stratified by estrus cycle (Di; n = 34 mice), (Pro; n = 13 mice), (Est; n = 16 mice), (Met; n = 6 mice); (C) Experiment was designed as in (A) except CD8α antibody was injected intravenously 12 hours after intracranial delivered peptides. Representative flow cytometry plots from brain tissue. Cells gated on CD8β +. Quantification of mean fluorescent intensity of IV-negative stained cells (gate indicated in histogram) with n = 4 for control and n = 4 for T_RM recall. All data are pooled from at least two independent experiments. Statistical significance was determined by unpaired two-tailed t test in all instances except the IgM panel in (B) where an unpaired two-tailed Mann–Whitney was performed (due to non-normal distribution of the data) where * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Error bars indicate standard error of mean. Di = diestrus; Est = estrus; FRT = female reproductive tract; LCMV = lymphocytic choriomeningitis virus; Ig = immunoglobulin; i.p. = intraperitoneal; IV = intravenous; Met = metestrus; Pro = proestrus; T_RM = Tissue-resident memory T cells; WT = wild-type.

Fig. 3

T_RM reactivation triggers rapid increase in antibody in the FRT. (A) Depo-Provera treated P14 chimeras were challenged with gp33 or control peptide transcervically and then 6 hours later, vaginal washes and FRT tissues were assayed for antibody by enzyme-linked immunosorbent assay; (B) Levels of antibody in vaginal washes 6 hours following T_RM reactivation; n = 12 mice for control and n = 14 mice for T_RM recall; (C) levels of antibody in FRT tissue 6 hours following T_RM reactivation; n = 10 mice for control and n = 12 mice for T_RM recall; (D) Representative IgM staining of FRT tissue where green = IgM and blue = DAPI stained nuclei. Scale bar = 500 μm; E: Quantification of D; n = 5 mice. All data are pooled from at least two independent experiments. Statistical significance was determined by unpaired two-tailed Mann–Whitney in (B) and (C) panel IgG2c, or by unpaired two-tailed t test in (E) and (C) panel IgG1 and IgG2b, where * p < 0.05, ** p < 0.01. Error bars indicate standard error of mean. DAPI = 4′,6-diamidino-2-phenylindole dihydrochloride; FRT = female reproductive tract; LCMV = lymphocytic choriomeningitis virus; Ig = immunoglobulin; T_RM = Tissue-resident memory T cells.

Fig. 4

T_RM reactivation triggers vascular permeability and neutralizing ab infiltration into the FRT. (A-C) Levels of albumin (n = 10 mice for control and n = 12 for T_RM recall) (A), intravenous transferred EBD (n = 8 mice) (B), and IV-transferred anti-OVA ab (n = 8 mice for control and n = 6 for T_RM recall) (C) in FRT tissue 6 hours following transcervical delivery of gp33 or control peptide to Depo-Provera treated P14 immune chimera mice. (D) Levels of PGDM1400 HIV neutralizing ab in the FRT after transferring 5 μm (left; n = 8 mice) or 50 μm (right; n = 8 mice) ab intravenous 2.5–3 hours prior to transcervical delivery of gp33 or control peptide; (E) Plaque reduction neutralization assay against VSV_OVA using FRT lysate from Depo-Provera treated P14 immune chimeras that were (right) or were not (left) previously infected with VSV_OVA. Data are 6 hours following transcervical delivery of gp33 or control peptide. Non-immune; n = 2 mice for control, experiment performed once, and n = 5 for T_RM recall. VSV immune; n = 7 mice for both groups. All data are pooled from at least two independent experiments unless noted. Statistical significance was determined by unpaired two-tailed t test in (A–C), unpaired two-tailed Mann–Whitney in (D), and a two-way analysis of variance with Sidak’s multiple comparisons test (E) where * p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001. Error bars indicate standard error of mean. ab = antibody; aOVA = anti-chicken ovalbumin; EBD = evans blue dye; FRT = female reproductive tract; T_RM = Tissue-resident memory T cells; VSV = virus expressing ova.