The development of inducible bronchus-associated lymphoid tissue depends on IL-17

Abstract

Ectopic or tertiary lymphoid tissues, such as inducible bronchus-associated lymphoid tissue (iBALT), form in nonlymphoid organs after local infection or inflammation. However, the initial events that promote this process remain unknown. Here we show that iBALT formed in mouse lungs as a consequence of pulmonary inflammation during the neonatal period. Although we found CD4(+)CD3(-) lymphoid tissue-inducer cells (LTi cells) in neonatal lungs, particularly after inflammation, iBALT was formed in mice that lacked LTi cells. Instead, we found that interleukin 17 (IL-17) produced by CD4(+) T cells was essential for the formation of iBALT. IL-17 acted by promoting lymphotoxin-α-independent expression of the chemokine CXCL13, which was important for follicle formation. Our results suggest that IL-17-producing T cells are critical for the development of ectopic lymphoid tissues.

Fig. 1. Preferential development of iBALT in neonatal mice

(a–c) Mice were intranasally administered LPS starting on day 2 after birth (a), 2 weeks after birth (b) or 3 weeks after birth (c) and were subsequently infected with influenza when they were 8 weeks old. Lungs were obtained 3 weeks after infection and frozen sections were probed with antibodies against B220, CD3 and CD21. Images are representative of at least 3 independent experiments with 4–8 mice per group. (d–e) Neonatal mice were intranasally administered PBS (d) or LPS (e) and then intranasally administered LPS when they were 8 weeks old. Lungs were obtained 3 weeks after the last LPS administration and frozen sections were analyzed by histology (Diff-Quick) or probed with antibodies against B220, Thy-1.2, PNAd or CD21. The boxes in histology panels indicate the areas analyzed by fluorescent microscopy in serial sections. Arrows indicate areas of iBALT. Images are representative of 2 experiments with 4–8 mice per group.

Fig. 2. iBALT forms independently of CCR2 and CCR6

(a) Neonatal C57BL/6 mice were intranasally administered either PBS or LPS. Lungs were collected 6 h after the last LPS administration and single cell suspensions were analyzed by flow cytometry for B220⁻CD11c⁺MHCII⁺CD11b^hiCD103⁻ DCs, B220⁻CD11c⁺MHCII⁺CD11b^loCD103⁺ DCs, CD11b⁺7/4⁺Ly6C⁺Ly6G⁺ neutrophils and CD11b⁺7/4⁺Ly6C⁺Ly6G⁻ monocytes. Data are representative of 3 experiments with 4–8 mice per group. (b) Neonatal C57BL/6, Ccr2^−/− and Ccr6^−/− mice were intranasally administered LPS. Lungs were obtained 1 week after the last LPS administration and frozen sections were probed with antibodies against B220, CD3 and CD21. Images are representative of 3 independent experiments with 3–5 mice per group. (c) The area of the lymphoid cell follicles in the lungs of mice of each genotype was determined using the outline tool in the Zeiss Axiovision software. Data are representative of 3 experiments with 3–5 mice per group and at least 1 slide per animal. Data are shown as mean ± standard deviation. (d) Neonatal mice were intranasally treated with LPS or PBS and total RNA was obtained from the lungs 6 h following the last LPS administration. RNA expression was analyzed by quantitative PCR for the indicated mRNAs and all values were normalized to GAPDH. These experiments were performed twice with 4–8 mice per litter. Data are shown as mean ± standard deviation.

Fig. 3. LTi cells are not required for iBALT development

(a) Neonatal mice were intranasally administered either PBS or LPS and single cell suspensions from lungs were analyzed by flow cytometry 1 week after the last LPS administration. Lineage⁺ cells were identified using a cocktail of antibodies against CD3, CD8, CD11b, CD11c, CD19, B220, NK1.1 and GR-1. The percentage of CD4⁺lineage⁻ cells is indicated in the boxes. (b) CD4⁺lineage⁻ LTi cells as well as CD4⁺CXCR5⁺lineage⁻ LTi cells were enumerated. The data in a and b are representative of 2 independent experiments with 4–8 mice per group. Dare are shown as mean ± standard deviation. (c) Neonatal mice were intranasally administered LPS and were subsequently infected with influenza when they were 8 weeks old. Lungs were obtained 3 weeks after infection and frozen sections were probed with antibodies against B220, Thy1.2, CD21 and PNAd. The images shown are representative of 4 experiments with 4–8 mice per group.

Fig. 4. Elevated expression of IL-17 in the lungs of LPS-treated neonates

Mice were intranasally administered 10 µg LPS 5 times (once every other day) starting on day 2, day 14 or 8 weeks after birth. Lungs were obtained 6 hours following the last LPS inoculation and total RNA was extracted. RNA expression was analyzed by quantitative PCR for the indicated mRNAs and all values were normalized to the expression of GAPDH. This experiment was performed 3 times with 4–8 mice per litter. Data are shown as mean ± standard deviation. Asterix indicate a significant difference (p<0.05 unpaired t test) from mice treated with LPS at 0 weeks.

Fig. 5. IL-17 is required for the development of iBALT

(a–e) Neonatal mice were intranasally administered LPS and lungs were obtained 1 week following the last LPS administration and probed with antibodies against B220 and CD21. Images are representative of at least three mice per group. (f) The area of the lymphoid clusters, B220⁺ B cell follicles or CD21⁺ FDC networks was determined using the outline tool in the Zeiss Axiovision software. Data are shown as mean ± standard deviation from multiple slides of at least 3 mice per group. Asterix indicate a significant difference (p<0.05 unpaired t test) from C57BL/6 mice. (g,h) Neonatal mice were intranasally administered LPS and lungs were obtained 6 hours following the last intranasal LPS administration and total RNA was extracted. RNA expression was analyzed by quantitative PCR for the indicated mRNAs and all values were normalized to GAPDH. Graphs indicate mean ± standard deviation. Asterix indicate a significant difference (p<0.05 unpaired t test) from C57BL/6 mice. These experiments were performed 2 times with 4–8 mice per litter. (i). Pulmonary fibroblasts were obtained from untreated neonatal lungs, stimulated with IL-17A or TNF and chemokine expression was measured by quantitative PCR. All values were normalized to GAPDH. Graphs indicate mean ± standard deviation. Asterix indicate a significant difference (p<0.05 unpaired t test) from C57BL/6 mice. This experiment was performed 3 times with similar results.

Fig. 6. IL-17 acts early in iBALT formation, but does not maintain its structure

(a) Neonatal mice were intranasally administered LPS and lungs were obtained at the indicated times and frozen sections were probed with antibodies against B220, CD3 and CD21. Images are representative of mice from 3 litters. (b) The total area of the CD21-CD35⁺ FDC networks in B cell follicles at each time was determined using the outline tool in the Zeiss Axiovision software. Bars represent mean ± standard deviation. This experiment was performed with at least 3 litters of mice with 4–8 mice per litter. (c) LPS was intranasally administered to neonatal mice and sLTβR-Ig or anti-IL-17 blocking reagents were intranasally administered either 1 day before the last LPS administration or 1 week after the last LPS administration. Lungs were obtained for histology 1 week following treatment. B cell clusters or FDC networks were enumerated by immunofluorescence microscopy. The area of individual B cell clusters or CD21⁺ FDC networks was determined using the outline tool in the Zeiss Axiovision software. Bars represent mean ± standard deviation obtained from multiple slides from at least 3 mice per group. Asterix indicate a significant difference (p<0.05 unpaired t test) from C57BL/6 mice.

Fig. 7. IL-17-producing T cells promote iBALT formation

(a) IL-17 expression was measured by quantitative PCR in αβT cells, γδT cells and CD4⁺ cells from lungs of LPS-treated neonates and in CD4⁺ cells from naïve LNs. Data are representative of 2 experiments from 5–10 pooled mice. (b–d) Frozen sections from LPS-treated neonatal lungs were probed with antibodies to IL-17A and CD4 (b) or with antibodies to CD3, CD11c, B220, NKp46 and γδTCR (c–d). Serial sections are shown. Images are representative of 2 experiments with 4–8 mice per litter. (e) Neonatal Tcrbd^−/− mice received no cells or CD4⁺ T cells and were intranasally administered LPS. Lungs were obtained 1 week following the last LPS administration and frozen sections were probed with antibodies to CD3, CD11c, B220, CD21 and PNA. (f–g) Rorc^−/− mice were depleted with anti-CD4 or isotype control one day before the last LPS administration. Lungs were obtained 1 week later and probed with antibodies to CD3, CD11c, B220, CD21 and PNA (f). Depletion of CD4⁺ αβT cells and γδT cells was confirmed using flow cytometry (g). Bars indicate mean ± standard deviation. These experiments were performed twice with 4–8 mice per litter. (h) γδT cells or αβT cells were adoptively transferred to neonatal, LPS-treated Tcrbd^−/− recipients. Lungs were obtained 1 week after the last LPS administration and frozen sections were probed with antibodies against CD21, CD3 and B220. Morphometric analysis of lymphoid areas was performed. Images are representative of 2 independent experiments with 4–8 mice per group. Graphs show the mean ± standard deviation of multiple slides from 4–8 mice per group. (i,j) OVA-specific T_FH, Th17 and T_FH17 cells were transferred to neonatal Tcrbd^−/− recipients, which were challenged with LPS and OVA and analyzed by immunofluorescence for CD21, CD3 and B220 1 week after the last OVA administration. Flow cytometry plots represent IL-17 and CXCR5 expression on cells prior to transfer. (j) The total area of B cell follicles was determined. Bars represent mean ± standard deviation. Asterix indicate a significant difference (p<0.05 unpaired t test) from the T_FH group. FACS plots, images and graphs represent 1 of 2 experiments with 3–7 mice per group.