Adenoviral vaccine induction of CD8+ T cell memory inflation: Impact of co-infection and infection order

Abstract

The efficacies of many new T cell vaccines rely on generating large populations of long-lived pathogen-specific effector memory CD8 T cells. However, it is now increasingly recognized that prior infection history impacts on the host immune response. Additionally, the order in which these infections are acquired could have a major effect. Exploiting the ability to generate large sustained effector memory (i.e. inflationary) T cell populations from murine cytomegalovirus (MCMV) and human Adenovirus-subtype (AdHu5) 5-beta-galactosidase (Ad-lacZ) vector, the impact of new infections on pre-existing memory and the capacity of the host's memory compartment to accommodate multiple inflationary populations from unrelated pathogens was investigated in a murine model. Simultaneous and sequential infections, first with MCMV followed by Ad-lacZ, generated inflationary populations towards both viruses with similar kinetics and magnitude to mono-infected groups. However, in Ad-lacZ immune mice, subsequent acute MCMV infection led to a rapid decline of the pre-existing Ad-LacZ-specific inflating population, associated with bystander activation of Fas-dependent apoptotic pathways. However, responses were maintained long-term and boosting with Ad-lacZ led to rapid re-expansion of the inflating population. These data indicate firstly that multiple specificities of inflating memory cells can be acquired at different times and stably co-exist. Some acute infections may also deplete pre-existing memory populations, thus revealing the importance of the order of infection acquisition. Importantly, immunization with an AdHu5 vector did not alter the size of the pre-existing memory. These phenomena are relevant to the development of adenoviral vectors as novel vaccination strategies for diverse infections and cancers. (241 words).

Fig 1. Simultaneous infection with MCMV and Ad-lacZ.

Groups of C57BL/6 mice were co-infected with a mixture of 2x10⁹ pfu Ad-LacZ and 5x10⁶ pfu MCMV by the i.v. route (N = 4). Groups of mice were infected with only MCMV or Ad-lacZ as control (N = 3). At the indicated time-points post-infection blood was sampled and the levels of Ad-lacZ-specific inflating memory (D8V, upper left) and central memory (I8V, upper right) and MCMV-specific inflating memory (M38, lower left) and central memory (M45, lower right) were measured by ex vivo tetramer staining.

Fig 2. Sequential infection with MCMV followed by Ad-LacZ.

(A) C57BL/6 mice were first infected with MCMV, then >50 days later were infected with Ad-lacZ. The size of the pre-existing MCMV specific M38 effector memory response after Ad-lacZ infection was measured in the blood and liver by ex-vivo tetramer staining. The percentage of pre-existing M38 response in the blood and (B) the absolute numbers in the liver after immunization with Ad-lacZ vector was determined at the indicated time points. (C) The kinetics and magnitude of the new inflating response (D8V) in groups of mice with latent MCMV was also measured by ex vivo tetramer staining compared against uninfected mice. The percentage of D8V tetramer+ cells in the blood and (D) the absolute numbers in the liver was measured at the indicated time points. The figures show the mean from 3–8 mice per time point obtained from 2 independent experiments. p values were measured by Mann-Whitney tests. *p<0.05.

Fig 3. Sequential infection with Ad-LacZ followed by MCMV.

C57BL/6 mice were first immunized with Ad-lacZ, then >50 days later were infected with MCMV. The size of the pre-existing Ad-lacZ specific M38 effector memory response after MCMV infection was measured in the blood and liver by ex-vivo tetramer staining. (A) The percentage of pre-existing inflating D8V tetramer+ population in the blood and (B) the absolute numbers in the liver after MCMV infection was determined at the indicated time points. (C) The kinetics and magnitude of the new inflating response (M38) in groups of mice previously immunized with Ad-lacZ was also measured by ex vivo tetramer staining and compared against uninfected mice. The percentage of M38 tetramer⁺ cells in the blood and (D) the absolute numbers in the liver was measured at the indicated time points. The figures show the mean from 3–8 mice per time point obtained from 2 independent experiments. p values were measured by Mann-Whitney tests followed by Dunn’s multiple comparison. *p<0.05, **p<0.005.

Fig 4. Effect of acute MCMV infection on pre-existing CD8⁺ D8V⁺ memory cells.

(A) Mice were first immunized with Ad-lacZ, then infected >50 days later with MCMV. Data are combined from two independent experiments (N = 7–9 mice per group). Peripheral lymphocytes were sampled at 21 days after MCMV infection and stained ex vivo with CD8, CD44 and CD62L. The proportions of effector memory (CD44⁺ CD62L low), central memory (CD44⁺CD62L⁺) and naïve (CD44low CD62L⁺) was determined. (B) Levels of D8V-specific tetramer population was measured between days 2–7 after infection with MCMV in the blood (N = 4–5 from two independent experiments) and at day 4 in the spleen, bone marrow, lymph nodes and liver. (C) In vivo CTL assay. Splenocytes from mice immunized with Ad-lacZ 45 days previously were isolated and labelled with CFSE then equal numbers of CFSE-labelled splenocytes were adoptively transferred into mice infected with MCMV at day 1 post-infection or a group of uninfected controls. The levels of transferred D8V⁺ effecter memory cells was followed in the blood over time by ex vivo tetramer staining (N = 6 per group from two independent experiments). To ensure equal numbers of splenocytes were transferred, the percentage of CFSE⁺CD8⁺ cells in naïve and MCMV-infected recipients were compared at the indicated time points. (D) The levels of the inflating effector memory D8V tetramer⁺ population was measured in naïve mice and a group acutely infected with MCMV at the indicated time points. p values were measured by Mann-Whitney test. *p<0.05.

Fig 5. Analysis of death pathways after MCMV infection.

(A) Microarray profiling of MCMV specific M38 CD8⁺ T cells and Ad-lacZ specific D8V CD8⁺ T cells was used for GSEA analysis for the KEGG apoptosis pathway comparing M38 inflationary cells at day 50 post infection to D8V inflationary cells at day 100 post infection. (B) Heatmap of the KEGG apoptosis pathway geneset from Fig 5A. Columns represent M38 or D8V samples and rows different upregulated (red) or downregulated (blue) genes. (C) A representative histogram of the levels of Fas on CD8 T cells isolated from bone marrow of Ad-lacZ only or Ad-lacZ-immune mice infected 4 days previously with MCMV. Levels of Fas on the surface of CD8 T cells (lower left) and D8V⁺CD8 T cells (lower right) isolated from the indicated organs at 4 days post-MCMV infection. (N = 4–5 per group from two independent experiments). (D) Single cell suspensions prepared from organs of mice at 4 days post-MCMV infection were cultured in vitro in complete RPMI for 2 days. The samples were gated on total CD8 T cells and the percentage of live CD8 T cells (left) and the percentage of D8V⁺ tetramer positive cells (right) within the live CD8 T cell population in each organ is shown. The data are from 4–5 mice per group from two independent experiments). (E) Mice were immunized with Ad-lacZ, then left for more than 55 days. One group of mice were injected i.v. with 100μg anti-FasL or isotype control antibody per mouse either one day before or on the same day as infection with 1x10⁶pfu MCMV. A second antibody treatment was performed 3 days post-MCMV. Blood was collected 9 days later and the levels of D8V⁺ CD8⁺T cells measured. Data from two independent experiments are shown. Each datapoint represents an individual mouse. p values determined by T-tests **p<0.005.

Fig 6. Effect of MCMV infection on pre-existing D8V responses in Ad-lacZ- immune IL-18KO mice.

(A) NK cell activity in IL-18 mice upon MCMV infection. The levels of activation markers NKG2ace, NKG2D, NK1.1 and KLRG1 (left to right) were measured by ex vivo staining of blood after MCMV infection (N = 8–12 mice per group from 2 independent experiments). (B) IL-18RKO mice were immunized with Ad-lacZ, then 50 days later were infected with MCMV (as indicated). The levels of pre-existing D8V⁺ CD8 T cells in the blood of IL-18R KO mice (N = 5 per group) were measured by ex vivo tetramer staining. p values were determined using one-way Anova. *p<0.05, **p<0.005, ***p<0.0005.

Fig 7. Effect of other infections on pre-existing Ad-lacZ memory responses.

(A) C57BL/6 mice were first immunized with Ad-lacZ and then 50 days later were infected with vaccinia i.p. The levels of pre-existing CD8⁺D8V⁺ in the blood after vaccinia infection was measured by ex vivo tetramer staining (N = 3 from one of two independent experiments). (B) Mice were immunized with Ad-lacZ, then >55 days later were infected with i.v. with 1x10³cfu/200μl Listeria-OVA. Blood was collected at the indicated timepoints, and the levels of CD8⁺D8V⁺ cells measured by surface staining. p values were obtained using T-tests * p<0.05.

Fig 8. Effect of boosting with Ad-lacZ on the depleted responses.

(A) C57BL/6 mice were first immunized with Ad-lacZ and then >50 days later were infected with MCMV i.v. After >50 days post- MCMV infection, the mice were boosted with a second dose of Ad-lacZ i.v. The levels of CD8⁺D8V⁺ in the blood was measured by ex vivo tetramer staining after primary Ad-lacZ infection. The data shown are from one of two independent experiments (N = 3). (B) The distribution of the boosted cells in non-lymphoid organs were measured at day 50 after 2^nd dose of Ad-LacZ. (C) The proportion of CD44⁺ CD62L^- expression in ex vivo peripheral blood 6 days after primary Ad-lacZ immunization, and 6 days after second dose of Ad-lacZ. (N = 4–6 mice from 2 independent experiments). (D) The figures show the proportion of CD44⁺CD62L^- (left column) and CD27^-CD62L^- (right column) expression in CD8⁺D8V⁺ cells from the blood (upper row) and liver (lower row) of Ad-lacZ only, Ad-lacZ+MCMV and Ad-lacZ+MCMV boosted with Ad-lacZ groups, as measured by ex-vivo staining. The data are from two or more independent experiments (N = 4–11). p values were measured by two-way Anova followed by Sidak’s multiple comparison test. *p<0.05, **p<0.005.