Distinct roles for neutrophils and dendritic cells in inflammation and autoimmunity in motheaten mice

Abstract

The motheaten mouse has long served as a paradigm for complex autoimmune and inflammatory disease. Null mutations in Ptpn6, which encodes the nonreceptor protein-tyrosine phosphatase Shp1, cause the motheaten phenotype. However, Shp1 regulates multiple signaling pathways in different hematopoietic cell types, so the cellular and molecular mechanism of autoimmunity and inflammation in the motheaten mouse has remained unclear. By using floxed Ptpn6 mice, we dissected the contribution of innate immune cells to the motheaten phenotype. Ptpn6 deletion in neutrophils resulted in cutaneous inflammation, but not autoimmunity, providing an animal model of human neutrophilic dermatoses. By contrast, dendritic cell deletion caused severe autoimmunity, without inflammation. Genetic and biochemical analysis showed that inflammation was caused by enhanced neutrophil integrin signaling through Src-family and Syk kinases, whereas autoimmunity resulted from exaggerated MyD88-dependent signaling in dendritic cells. Our data demonstrate that disruption of distinct Shp1-regulated pathways in different cell types combine to cause motheaten disease.

Figure 1. Disease phenotype in Ptpn6^fl/flS100a8-cre and Ptpn6^fl/flItgax-cre mice

(A, C) Mice of the indicated genotypes were monitored every two (Ptpn6^fl/flS100a8-cre) or four (Ptpn6^fl/flItgax-cre) weeks for paw inflammation or lymphadenopathy, respectively. The presence of these phenotypes was scored, and the percentage of “disease-free” mice at each time point was graphed using Prism. (B) H&E-stained sections of normal and inflamed paws from wild type S100a8-cre or Ptpn6^fl/flS100a8-cre mice. Original magnification ×10 (top) and ×40 (bottom). Thickened epidermal and dermal layers are indicated by blue and black arrowheads, respectively; bone marrow hypercellularity is indicated by the asterisk. (D) Lymph node weights, relative to total mouse weight, for the indicated genotypes are shown; each circle represents a single mouse. ***p <0.001. Photograph shows representative lymph nodes from wild type Itgax-cre or Ptpn6^fl/flItgax-cre mice. Scale bars, 5mm. (E) Spleen weights, relative to total mouse weight, for the indicated genotypes are shown; each circle represents a single mouse. *p<0.05, ***p <0.001. Photograph shows representative spleens from wild type Itgax-cre or Ptpn6^fl/flItgax-cre mice. Scale bars, 5mm. (F) Detection of anti-nuclear antibodies in serum by immunostaining of HEp-2 cells. (G) Detection of anti-DNA IgG antibodies in serum by ELISA (error bars indicate s.e.m.) **p<0.01. (H) H&E-stained sections of kidneys, showing glomerulonephritis in Ptpn6^fl/flItgax-cre mice. Original magnification ×20. Data were pooled from 6 independent experiments. See also Figure S1.

Figure 2. Lymphoid and DC expansion and activation in Ptpn6^fl/flItgax-cre mice

Cells were harvested from spleens and bone marrow, counted by using a Nucleocounter™, stained with fluorescently conjugated antibodies and analyzed by flow cytometry. (A) Total numbers of splenocytes in Ptpn6^fl/flS100a8-cre and Ptpn6^fl/flItgax-cre mice, compared with controls at 15–20 weeks of age; each circle represents one mouse. (B) Total numbers of CD11b⁺Gr1^hi neutrophils in the spleen and bone marrow of Ptpn6^fl/flS100a8-cre mice. (C) H&E-stained sections of spleens from 40 week-old Ptpn6^+/+Itgax-cre and Ptpn6^fl/flItgax-cre mice. Original magnification ×4 (top) and ×10 (bottom). (D) Total numbers of splenic neutrophils and monocyte and macrophage subsets in Ptpn6^fl/flItgax-cre mice. (E) Total numbers of splenic T and B cells in Ptpn6^fl/flS100a8-cre and Ptpn6^fl/flItgax-cre mice, compared with controls. (F, G) Numbers of total splenic cDCs (CD11c^hiMHCII^hi) or pDCs (CD11c^intB220⁺Ly6c⁺) and expression of activation markers in Ptpn6^fl/flItgax-cre mice. Each bar represents an n of 5 – 10 mice at 15–20 weeks of age per group (error bars indicate s.e.m.) *p<0.05, **p<0.01, ***p <0.001. Data were pooled from 6 independent experiments. See also Figure S2.

Figure 3. Neutrophils from Ptpn6^fl/flS100a8-cre mice show hyperactive integrin signaling

(A) Mice of the indicated genotypes were monitored weekly for paw inflammation, and the percentage remaining “disease-free” was graphed using Prism. (B) Neutrophils isolated from mice of the indicated genotypes were plated onto the integrin ligand poly-RGD (pRGD) in the presence or absence of TNF-α, and superoxide release was measured. (C, D) Neutrophils from Ptpn6^fl/fl or Ptpn6^fl/flS100a8-cre mice were kept in suspension or plated onto pRGD-coated plates for the indicated times, lysed and analyzed by SDS-PAGE, followed by immunoblotting with the indicated antibodies. Data shown are representative of at least 3 individual experiments (error bars indicate s.e.m.). See also Figure S3.

Figure 4. Neutrophil-specific Syk deficiency reverses the inflammatory disease in Ptpn6^fl/flS100a8-cre mice

(A) Mice of the indicated genotypes (wild type C57BL/6, n=14; Ptpn6^fl/flS100a8-cre, n=17; Ptpn6^fl/flSyk^fl/flS100a8-cre, n=9) were monitored for paw inflammation, and the percentage of disease-free mice at each time point was graphed using Prism, ***p <0.001. (B) Spleen weights, relative to total mouse weight, for the indicated genotypes; each circle represents a single mouse. (C–E) Spleen and bone marrow cells from mice of the indicated genotypes were harvested, counted using a nucleocounter™, then stained with fluorescently conjugated antibodies and analyzed by flow cytometry. Each circle represents a single mouse. Bars represent the average of 9–17 mice per group (error bars indicate s.e.m.) *p<0.05, **p<0.01, ***p <0.001. (F) Neutrophils isolated from mice of the indicated genotypes were plated onto the integrin ligand pRGD in the presence or absence of TNF-α, and superoxide release was measured (error bars indicate s.e.m.). (G) Neutrophils from wild type C57BL/6, Ptpn6^fl/flS100a8-cre and Ptpn6^fl/flSyk^fl/flS100a8-cre mice were kept in suspension or plated onto pRGD-coated plates for 15 mins, lysed and analyzed by SDS-PAGE, followed by immunoblotting with the indicated antibodies. Data shown are representative of at least 3 individual experiments.

Figure 5. Deletion of Myd88 in DCs reverses the autoimmunity in Ptpn6^fl/flItgax-cre mice

(A, B) Spleen and lymph node weights were obtained from the indicated groups of mice at 40 weeks of age and expressed as a fraction of total mouse weight. Each circle represents a single mouse, ***p <0.001. (C) Detection of anti-nuclear antibodies in serum by immunostaining of HEp-2 cells. (D) Detection of anti-DNA IgG antibodies in serum by ELISA (error bars indicate s.e.m.) *p<0.05, **p<0.01. (E–H) Cells were harvested from the spleens of mice with the indicated genotypes, counted by using a Nucleocounter™, stained with fluorescently conjugated antibodies and analyzed by flow cytometry. Each bar represents data from 5–9 mice (error bars indicate s.e.m.) *p<0.05, **p<0.01, ***p <0.001. See also Figure S4.

Figure 6. Splenic DCs from Ptpn6^fl/flItgax-cre mice show exaggerated TLR signaling

(A) BMDCs from mice of the indicated genotypes were stimulated with 100 ng/ml LPS for the indicated times, lysed and analyzed by SDS-PAGE followed by immunoblotting with the indicated antibodies. Data shown are representative of at least 5 individual experiments. (B) Splenocytes harvested from mice of the indicated genotypes were stimulated with 1 µg/ml LPS or 2 µg/ml CpG for the indicated times. Cells were then fixed with formaldehyde, permeabilized with methanol, stained with fluorescently conjugated anti-phospho-Erk, and analyzed by flow cytometry. The histograms represent the CD11c^hi DC gate. (C) Splenocytes were harvested from mice of the indicated genotypes, treated with brefeldin A, and stimulated with 1 µg/ml LPS or 2 µg/ml CpG for 5 hours. Cells were fixed and stained with fluorescently labeled anti-TNF-α and anti-CD11c, and analyzed by flow cytometry. Data are representative of 3 independent experiments done on different cohorts of mice. See also Figure S5.

Figure 7. Deletion of Myd88 in DCs reverses the exaggerated TLR signaling in Ptpn6^fl/flItgax-cre mice

(A) Splenocytes were harvested from mice of the indicated genotypes, and then stimulated with or without 1 µg/ml LPS for 30 mins. Cells were fixed with formaldehyde, permeabilized with methanol, and stained with fluorescently conjugated anti-phospho-Erk, followed by flow cytometry. The histograms represent CD11c^hi cells. (B) Splenocytes were harvested from mice with the indicated genotypes, then treated with brefeldin A and stimulated with 1 µg/ml LPS for 5 hours. Cells were fixed, stained with fluorescently conjugated antibodies, including anti-TNF-α, and analyzed by flow cytometry. Dot plots show CD11c^hi cells. Data are representative of 3 independent experiments.