Induction of memory-like dendritic cell responses in vivo

Abstract

Dendritic cells (DCs), a vital component of the innate immune system, are considered to lack antigen specificity and be devoid of immunological memory. Strategies that can induce memory-like responses from innate cells can be utilized to elicit protective immunity in immune deficient persons. Here we utilize an experimental immunization strategy to modulate DC inflammatory and memory-like responses against an opportunistic fungal pathogen that causes significant disease in immunocompromised individuals. Our results show that DCs isolated from protectively immunized mice exhibit enhanced transcriptional activation of interferon and immune signaling pathways. We also show long-term memory-like cytokine responses upon subsequent challenge with the fungal pathogen that are abrogated with inhibitors of specific histone modifications. Altogether, our study demonstrates that immunization strategies can be designed to elicit memory-like DC responses against infectious disease.

Fig. 1

Pulmonary infection with C. neoformans strain H99γ accelerates cDC1 response. BALB/c mice were inoculated via intranasal inhalation with 10⁴ CFU C. neoformans strain H99γ or H99. Pulmonary leukocytes were then isolated from the right lobes of enzymatically digested lungs on days 7, 14, and 20 post-inoculation and analyzed by flow cytometry. Individual mice were weighed every 72 h until conclusion of the observation period (day 20 post-inoculation). Pulmonary, brain, and spleen fungal burden was quantified from tissue homogenates on days 7, 14, and 20 post-inoculation. a Absolute numbers of CD103⁺ conventional (c) DCs (CD11c⁺/CD24⁺/CD11b⁻/CD103⁺), and b CD11b⁺cDCs (CD11c⁺/CD24⁺/CD11b⁺/CD172α⁺) in the lung. Data shown are from three experiments for days 7 and 14 and two experiments for day 20 with five mice per group. c Data shown are from 10 mice per group. d–f Pulmonary, brain, and spleen fungal burden data shown are from three experiments with 4–5 mice per group and results expressed as mean log CFU per gram of tissue. Bars indicate the mean ± standard error of the mean (SEM). (*p < 0.05, **p < 0.01, ***p < 0.001); unpaired Student’s t-test (two-tailed)

Fig. 2

C. neoformans strain H99 or H99γ infection results in differential DC phenotypes. BALB/c mice were inoculated via intranasal inhalation with 10⁴ CFU C. neoformans strain H99γ or H99. a, b Pulmonary leukocytes were isolated on days 5, 7, and 14 post-inoculation and depleted of CD3⁺ cells by positive selection using α-CD3e labeled magnetic beads and macrophages by positive selection using biotinylated α-F4/80 and α-biotin magnetic beads. Real-time PCR analysis of each transcript was normalized to GAPDH. Bars represent the log10-fold change in gene expression during infection with C. neoformans strain H99γ compared to wild-type C. neoformans strain H99. c NOS2:Arg1 expression ratio. d CXCL9:FIZZ1 expression ratio. Data shown are cumulative of three independent experiments utilizing pooled DCs from five mice per group per experiment. Bars indicate the mean ± standard error of the mean (SEM). (*p < 0.05, **p < 0.01, ***p < 0.001); unpaired Student’s t-test (two-tailed)

Fig. 3

Protective immunization induces pro-inflammatory DC phenotype during challenge. BALB/c mice were intranasally immunized with 10⁴ CFU of C. neoformans strain H99γ or HKH99γ in 50 µl of sterile PBS. Seventy days later, mice were left unchallenged (UC) or subsequently infected with 10⁴ CFU of C. neoformans strain H99. Pulmonary leukocytes were isolated from lung tissues by enzymatic digestion on days 1 and 3 post-inoculation or from immunized but unchallenged mice. Leukocyte populations were depleted of CD3⁺ cells by positive selection using α-CD3e labeled magnetic beads and macrophages by positive selection using biotinylated α-F4/80 and α-biotin magnetic beads. DCs were isolated by positive selection using CD11c⁺ magnetic beads. a, b Real-time PCR analysis of each transcript was normalized to GAPDH. Bars represent the log₁₀-fold change in gene expression in DCs from H99γ-immunized mice compared to non-protected mice. c NOS2:Arg1 expression ratio. d CXCL9:FIZZ1 expression ratio. Data shown are cumulative of three independent experiments utilizing pooled DCs from 20–25 mice per group per experiment and bars indicate the mean ± standard error of the mean (SEM). (*p < 0.05, **p < 0.001); unpaired Student’s t-test (two-tailed)

Fig. 4

Enhanced interferon gamma response networks in DCs from protectively immunized mice. BALB/c mice were immunized via intranasal inhalation with 10⁴ CFU of C. neoformans strain H99γ or HKH99γ in 50 µl of sterile PBS. Seventy days later, mice were challenged with 10⁴ CFU of C. neoformans strain H99 via intranasal inhalation. Pulmonary leukocytes were isolated from lung tissues by enzymatic digestion on day 1 post-challenge. Leukocyte populations were depleted of CD3⁺ cells by positive selection using α-CD3e labeled magnetic beads and macrophages by positive selection using biotinylated α-F4/80 and α-biotin magnetic beads. DCs were isolated by positive selection using CD11c⁺ magnetic beads. Total mRNA from isolated DC populations was extracted and RNA-seq was performed. a Differently expressed genes and top canonical pathways and networks predicted by the Ingenuity Pathway Analysis software protectively immunized mice compared to non-protectively immunized mice as ranked by p-value. b Top networks up-regulated in day 1 post-challenge pulmonary DCs from protectively immunized mice compared to non-protectively immunized mice. The red color in the network figure corresponds to the log2 of expression fold change and represents increased gene expression in DCs from protectively compared to non-protectively immunized mice. c Fold change values for top 12 up-regulated genes from H99γ immunized DCs day 1 post-challenge. d Top six GO terms from H99γ immunized DCs day 1 post-challenge. Data were generated from a merged data set from three independent experiments with 20–25 mice per group

Fig. 5

DCs from protectively immunized mice exhibit enhanced cytokine recall responses. a BALB/c mice were intranasally immunized with 10⁴ CFU of C. neoformans strain H99γ or HKH99γ in 50 µl of sterile PBS. After 70 days, DCs were isolated from the spleens of naïve, HKH99γ immunized or H99γ immunized mice after depletion of CD3⁺ and F4/80⁺ cells and stimulated with different antigens. b–e Isolated DCs were cultured alone or in the presence of a calcineurin alpha 1 subunit mutant (cna1Δ) of H99 in duplicate for 24 h after which supernatants were collected for cytokine analysis. f–i Isolated DCs were cultured in the presence of C. neoformans cna1Δ, LPS, heat-killed Staphylococcus aureus, or heat-killed Candida albicans yeast for 24 h and cytokine levels from the supernatants were analyzed. Results are cumulative of three experiments utilizing pooled DCs from 20 mice per group. Bars indicate the mean ± standard error of the mean (SEM). Symbols where significant differences were observed: a = compared to naïve, b = compared to HKH99γ, * = compared to media alone for that immunization group (*p < 0.05, **p < 0.01, ***p < 0.001); one-way ANOVA with the Tukey’s multiple comparison test

Fig. 6

DC cytokine recall responses in the presence of histone modification inhibitors. a BALB/c mice were immunized via intranasal inhalation with 10⁴ CFU of C. neoformans strain H99γ or HKH99γ in 50 µl of sterile PBS. b–d After 70 days, DCs were isolated from the spleens of naïve, HKH99γ-immunized or H99γ-immunized mice after depletion of CD3⁺ and F4/80⁺ cells. Isolated DCs were cultured in media alone or in media containing live cna1Δ ± MTA (methyltransferase inhibitor) or pargyline (demethylase inhibitor) for 24 h and cytokine levels from the supernatants were analyzed. e Isolated DCs were also cultured in media alone or in media containing live cna1Δ ± GSK343 (EZH2 inhibitor that trimethylates H3K27), UNC0638 (G9a inhibitor that methylates H3K9) or MI-2 (MLL inhibitor that trimethylates H3K4) for 24 h and IL-2 production in culture analyzed. Results are cumulative of three experiments utilizing pooled DCs from 20 mice per group. Bars indicate the mean ± standard error of the mean (SEM). Asterisks indicate where significant differences were observed compared to DCs cultured in media containing live cna1Δ (*p < 0.05), one-way ANOVA with the Tukey’s multiple comparison test