BP180 (also termed type XVII collagen) is a hemidesmosomal protein and plays a critical role in cell–cell matrix adhesion in the skin; however, its other biological functions are largely unclear. In this study, we generated a BP180 functional–deficient mouse strain by deleting its extracellular domain of humanized NC16A (termed ΔNC16A mice). We found that BP180 is expressed by bone marrow mesenchymal stem cells (BM-MSC), and its functional deficiency leads to myeloid hyperplasia. Altered granulopoiesis in ΔNC16A mice is through bone marrow stromal cells evidenced by bone marrow transplantation. Furthermore, the level of G-CSF in bone marrow and circulation were significantly increased in ΔNC16A mice as compared with wild-type mice. The increased G-CSF was accompanied by an increased activation of the NF-κB signaling pathway in bone marrow and BM-MSC of ΔNC16A mice. Blockade of G-CSF restored normal granulopoiesis in ΔNC16A mice. Inhibition of NF-κB signaling pathway significantly reduces the release of G-CSF from ΔNC16A BM-MSC in vitro and the level of serum G-CSF in ΔNC16A mice. To our knowledge, these findings provide the first direct evidence that BP180 plays an important role in granulopoiesis through regulating NF-κB signaling pathway in BM-MSC.

Also termed a type XVII collagen, BP180 is a 180-kDa transmembrane hemidesmosomal protein (1). The intracellular region of BP180 is linked to the intermediate filament network, and its extracellular portion is anchored into the basement membrane zone through interaction with extracellular matrix proteins (24). BP180 as a critical cell–cell matrix adhesion protein was illustrated by two human subepidermal blistering diseases: bullous pemphigoid (BP), caused by anti-BP180 autoantibodies that attack and impair function of BP180 autoantigen in basal keratinocytes (5, 6); and generalized atrophic benign epidermolysis bullosa (a type of junctional epidermolysis bullosa [JEB]), caused by BP180 gene mutations (79). However, whether BP180 is involved in other biological processes and pathological conditions is largely unknown.

BP is the most common skin autoimmune blistering disease, characterized by autoantibodies directed against the hemidesmosomal proteins BP230 and BP180, dermal–epidermal junction separation and inflammatory cell infiltration (10). Patients with BP have eosinophils and neutrophils in the bullous cavity and dermis of the lesioned skin (1). The membrane-proximal noncollagen (NC) 16A linker domain harbors multiple epitopes recognized by BP autoantibodies (11, 12). Although human BP180 shares high overall homology with mouse BP180, the NC16A domain is very poorly conserved with the counterpart of mouse BP180 (termed NC14A), resulting in a lack of immune cross-reactivity between these two species (6). Animal models of BP demonstrated that anti-NC14A and anti-NC16A Abs induce subepidermal blistering in mice associated with neutrophil infiltration (6, 13).

JEB is a group of rare genetic diseases characterized by marked skin fragility and blister formation following minor trauma, along with eosinophilia, eosinophils, and/or neutrophils infiltration in the dermis (14).

BP180 mutations lead to the partial or complete loss of BP180 function, causing JEB, which is commonly termed as non-Herlitz JEB (15, 16). Patients with non-Herlitz JEB exhibited a wide spectrum of clinical outcomes, from mild to severe phenotypes of JEB symptoms (17). In general, the milder forms of JEB are associated with missense or splice-site mutations and the presence of truncated BP180 protein in the skin (1719). Previous reports demonstrated that deletion of NC14A domain of mouse BP180 leads to loss of cell–cell matrix adhesion and granulocyte infiltration in the skin (20). However, the mechanisms underlying the increased granulocytes in BP/JEB patients and mice lacking BP180 function are unknown.

Multipotent hematopoietic stem cells undergo intermediate development stages, giving rise to an linIL7RKIT+Sca-1CD34+FcrRlo common myeloid progenitor, and common myeloid progenitors then differentiate into megakaryocyte/erythrocyte and granulocyte/macrophage progenitors (2124). Neutrophils, eosinophils, and basophils constitute the three main types of granulocytes. Large quantities of neutrophils are produced and stored in the bone marrow (25, 26). Granulopoiesis is regulated by cell-autonomous and non–cell-autonomous mechanisms (27, 28). G-CSF is a major determinant of non–cell-autonomous mechanism in granulocyte differentiation, proliferation, and maturation (27, 29, 30). G-CSF is also important for neutrophil mobilization (31). G-CSF is secreted by bone marrow stromal cells, endothelial cells, macrophages, and fibroblast cells (32). Deficiency of G-CSF or G-CSFR causes neutropenia in mice, and administration of G-CSF is a clinical practice to treat neutropenia (33), demonstrating that G-CSF is an important extrinsic regulator for neutrophils (32, 34).

Homeostasis of hematopoietic stem cells in bone marrow is dependent on multiple signaling pathways (35, 36), including Notch and NF-κB as two of the most investigated signaling pathways (37, 38). The bone marrow niche regulates hematopoiesis in an NF-κB–dependent manner (3941). NF-κB is a protein complex composed of inhibitory component IκBα and DNA binding p65/39. Upon external signal stimulations, NF-κB is phosphorylated. As a result, phosphorylated IκBα is degraded and phosphorylated p65 is transported to nucleus to activate the transcription of its downstream genes (42), including the gene encoding G-CSF (43). Loss of transcriptional repression of NF-κB activation in IκBα-deficient mice leads to an uncontrolled expansion of myeloid cells via G-CSF (44, 45).

In this study, we used mouse strains lacking BP180 function by deleting the humanized NC16A domain of the molecule to investigate new functions of BP180 (46). We show that deletion of the BP180 NC16A domain leads to altered granulopoiesis. To our knowledge, this is the first report linking BP180 in bone marrow stromal cells and granulopoiesis.

Humanized NC16A and ΔNC16A mice were described previously (13, 46). Briefly, mouse NC14A-encoding exons were replaced by human NC16A-encoding exons flanked by loxP sites. This approach offers one step transgenic manipulation to produce both humanized NC16A mice and conditional NC16A-deletion mice. NC16A mice were then bred with germline Cre mice (The Jackson Laboratory) (both on C57BL/6J background), leading to removal of NC16A-encoding exons and expression of NC16A-truncated BP180 in ΔNC16A mice (46). Mice with skin-specific deletion of NC16A were generated by crossing NC16A mice with K14Cre mice (46). C57BL/6 (CD45.1 and CD45.2) mice were purchased from National Cancer Institute. To make mixed bone marrow chimera, bone marrow from ΔNC16A CD45.2 and wild-type (WT) CD45.1 mice were mixed at 1:1 ratio, and then 5 × 106 cells were transplanted to sublethal irradiated (700 cGy) WT CD45.1.2 recipient mice by i.v. administration (47, 48).

All the mice were bred and hosted under specific pathogen-free conditions in the animal facilities of the University of North Carolina at Chapel Hill. Animal care and animal experiments were conducted in accordance with the Animal Care Committee at University of North Carolina at Chapel Hill.

Bone marrow cell isolation and culture were as described (49, 50). Briefly, bone marrow cell suspensions were isolated from 8 to 12 wk ΔNC16A and WT control mice by flushing the marrow from femurs a few times with 10 ml PBS/2% FBS in syringe with a 22-G needle. The cells were centrifuged for 5 min at 1000 rpm, 4°C after RBCs were lysed and removed. Cells were counted and seeded with mesenchymal stem cell (MSC) culture medium (MesenCult MSC culture medium; STEMCELL Technologies, Vancouver, BC, Canada) according to the protocol provided by the manufacture. Only the adherent monolayer cells were used as MSC after two washes with PBS to remove all nonadherent cells.

The colony-forming capacity of bone marrow cells was estimated in Mouse Methylcellulose Complete Media containing mouse IL-3, IL-6, stem cell factor, and human insulin, according to the manufacturer’s instructions (Mouse Methylcellulose Complete Media without Epo; R&D Systems) (51). Briefly, 3 × 104 bone marrow mononuclear cells were mixed well in 1 ml media, plated in 3-cm diameter culture dishes, and then incubated at 37°C in a 5% CO2 incubator. At 9 d after plating, the total number of colonies in each dish was counted under a microscopy.

Bone marrow, spleen, lymph nodes, and peripheral blood were taken from mice. A total of 5 × 106 cells were used for staining. Bone marrow single-cell suspension was made by flushing femurs a few times by 22-gauge needle and 10-ml syringe with 10 ml PBS/2% FBS. Spleen and lymph nodes were mashed by slides carefully and then passing through 70-μm cell strainer (Falcon; Corning) to isolate single-cell suspension. All cell suspensions were treated with ammonium–chloride–potassium buffer for RBC lysis. A total of 5 × 106 cells were used for staining. Anti–Gr-1 and anti-CD11b Abs were used to detect granulocytes. Anti-CD3 Ab was used to detect T cells. Anti-CD19 Ab was used to detect B cells. Anti-CD127, -CD34, -CD16, –Sca-1, and –c-Kit Abs were used to identify different population of granulocyte progenitors (Fig. 3A). These Abs were purchased from BioLegend (San Diego, CA).

Expression of BP180 in different tissues was determined by immunoblotting and immunofluorescence using standard techniques as previously described (13). The anti-NC16A Ab was from BP patient serum, whereas rabbit anti–NC1-3 polyclonal Ab was made in-house. Femurs were fixed in 4% paraformaldehyde overnight and then decalcification by 14% EDTA overnight and then tissue was rinsed by PBS three times. The femurs were fixed in OCT (Tissue-TEK). Cryosections of the OCT-embedded femurs were stained and were imaged by a Zeiss LSM 700 Confocal Laser Scanning Microscope and ZEN imaging software. The pictures were analyzed by ImageJ software. Bone marrow cells and bone marrow–derived MSC (1 × 106) were used for NF-κB signaling pathway determination. After Pierce BCA protein assay (Thermo Fisher Scientific), proteins were probed by the following Abs: anti–phospho-ErK, anti-ErK, anti-IκBα, anti–phospho-IκBα, anti-p65, anti–phospho-p65, and anti-rabbit secondary Ab (Cell Signaling Technology).

Levels of G-CSF and GM-CSF in bone marrow supernatant and serum samples from ΔNC16A, K14Cre/ΔNC16A, and WT control mice were quantified by ELISA kits (Abcam, MA). ELISA Max Mouse GM-CSF ELISA kits were purchased from BioLegend. To collect bone marrow supernatants, marrows from femurs were flushed out with PBS and centrifuged. Collected supernatants were stored in −80°C before use.

The data are expressed as mean ± SEM and were analyzed using the Student t test. A p value <0.05 was considered significant.

Previously, we generated humanized NC16A mice to investigate pathogenesis of BP and ΔNC16A mice for studying the role of BP180 in skin inflammation (13, 46). ΔNC16A mice showed no clinical skin phenotypes after birth but began to develop minor skin lesions at the age of 8–12 wk and erosions mostly on snout and ears at 16 wk old (Supplemental Fig. 1A). ΔNC16A mice also showed retarded growth (Supplemental Fig. 1B). Genotyping of mouse tail DNA confirmed the lack of exons 18 and 19 in ΔNC16A mice (data not shown). Interestingly, necropsy on 8–12-wk-old ΔNC16A mice revealed splenomegaly and bone marrow color change from pink to gray in ΔNC16A mice (Fig. 1A). Spleen/body weight ratio was significantly increased in ΔNC16A compared with WT control mice (Supplemental Fig. 1C).

FIGURE 1.

Granulocyte hyperplasia in ΔNC16A mice. (A) Biopsy examination of spleen and bone. Representatives of gross anatomy of spleen and bone marrow of WT and ΔNC16A mice. Splenomegaly was observed in ΔNC16A mice at age of 8–12 wk, and the bone of ΔNC16A mice was less red than WT mice. (B) Granulocyte infiltration. H/E staining showed increased granulocytes in blood, increased cellularity in bone marrow (BM), increased red pulp in spleen, and increased infiltrating granulocytes in dermis of ΔNC16A mice compared with WT mice. Original magnification ×100. (C) Hematological examination. Complete blood count showed significantly increased WBCs, granulocytes, and monocytes in peripheral blood of ΔNC16A mice compared with WT mice (n = 8). (D) Relative cell population. Percentage of granulocyte populations was also increased significantly, whereas lymphocyte population was significantly decreased in peripheral blood of ΔNC16A mice compared with WT mice (n = 8). (E) Granulocyte cell frequency. Granulocytes in BM, spleen, and blood of WT and ΔNC16A mice were analyzed by flow cytometry using forward scatter (cell size) and side scatter (granularity). Granulocyte populations in these three immune sites were around three folds increased in ΔNC16A mice than WT mice, and the increase was statistically significant (n = 6). (F) Granulocyte cell number. Neutrophil numbers were quantified by flow cytometry using neutrophil specific surface markers CD11b and Gr-1. Neutrophils were significantly increased in ΔNC16A mice than WT mice, and the difference is statistically significant (n = 6). **p < 0.01, ***p < 0.001, ***p < 0.001. Lym, lymphocyte; Mono, monocyte.

FIGURE 1.

Granulocyte hyperplasia in ΔNC16A mice. (A) Biopsy examination of spleen and bone. Representatives of gross anatomy of spleen and bone marrow of WT and ΔNC16A mice. Splenomegaly was observed in ΔNC16A mice at age of 8–12 wk, and the bone of ΔNC16A mice was less red than WT mice. (B) Granulocyte infiltration. H/E staining showed increased granulocytes in blood, increased cellularity in bone marrow (BM), increased red pulp in spleen, and increased infiltrating granulocytes in dermis of ΔNC16A mice compared with WT mice. Original magnification ×100. (C) Hematological examination. Complete blood count showed significantly increased WBCs, granulocytes, and monocytes in peripheral blood of ΔNC16A mice compared with WT mice (n = 8). (D) Relative cell population. Percentage of granulocyte populations was also increased significantly, whereas lymphocyte population was significantly decreased in peripheral blood of ΔNC16A mice compared with WT mice (n = 8). (E) Granulocyte cell frequency. Granulocytes in BM, spleen, and blood of WT and ΔNC16A mice were analyzed by flow cytometry using forward scatter (cell size) and side scatter (granularity). Granulocyte populations in these three immune sites were around three folds increased in ΔNC16A mice than WT mice, and the increase was statistically significant (n = 6). (F) Granulocyte cell number. Neutrophil numbers were quantified by flow cytometry using neutrophil specific surface markers CD11b and Gr-1. Neutrophils were significantly increased in ΔNC16A mice than WT mice, and the difference is statistically significant (n = 6). **p < 0.01, ***p < 0.001, ***p < 0.001. Lym, lymphocyte; Mono, monocyte.

Close modal

To explain these gross alterations in spleen and bone marrow, we analyzed the hematopoiesis system of ΔNC16A mice by routine histology, hematological examination, and flow cytometry. Histological examination showed increased granulocytes in blood and dermis, increased cellularity in bone marrow, and increased red pulp in spleen of ΔNC16A mice as compared with WT control (Fig. 1B). Hematological test showed significantly increased total cell number (Fig. 1C) and percentage (Fig. 1D) of granulocytes in circulation in ΔNC16A compared with WT mice. Normal granulocyte number range of WT mice is <20% in WBCs (41), whereas ΔNC16A mice manifested an abnormally high percentage of granulocytes in WBCs in the peripheral blood. Flow cytometry analysis also showed significantly increased granulocytes in bone marrow, spleen, and peripheral blood in ΔNC16A mice as compared with WT controls using forward scatter (cell size) and side scatter (granularity) (Fig. 1E) and granulocyte number counts based on the surface markers CD11b and Gr-1 (Fig. 1F). These results demonstrated that a lack of BP180 function caused myeloid hyperplasia.

In bone marrow hematopoietic stem cells are first differentiated into common myeloid progenitor and granulocyte/monocyte progenitor and then through various developmental stages, eventually differentiated to neutrophils, basophils, and eosinophils (Fig. 2A) (42, 52, 53). Because there is granulocyte hyperplasia in ΔNC16A mice, we determined whether the progenitors of granulocytes in bone marrow of ΔNC16A mice were changed by flow cytometry using LinIL-7RSCA-1c-kit+CD34+ CD16+ surface markers combination panel (Fig. 2A) (43, 44). We found that there was more than a 2-fold increase in granulocyte/macrophage progenitor population in ΔNC16A mice compared with WT control (Fig. 2B), which was statistically significant (Fig. 2C). We also observed a significantly higher number of CFUs of granulocytes/monocytes from ΔNC16A bone marrow than WT control mice (Fig. 2D). The hemoglobin content and platelet counts were not significantly different between WT and ΔNC16A mice (Supplemental Fig. 2A, 2B). These results demonstrate that a lack of BP180 leads to myeloid progenitor lineage skewing toward granulocyte/macrophage progenitors, resulting in increased granulocyte production.

FIGURE 2.

Increased granulocyte/monocyte progenitor in ΔNC16A mice. (A) Diagram showing different stages of granulopoiesis and surface markers expressed in different developmental stages. HSC, hematopoietic stem cell; MEP, megakaryocyte/erythrocyte progenitor; MMP, multipotent progenitor. (B) Flow cytometry. Bone marrow cells were gated on Lin and IL-7R double-negative cells, and c-Kit–positive and Sca-1–negative cell population was used for further analysis. CD34 and CD16 were stained to distinguish granulocyte/macrophage progenitor, common myeloid progenitor, and MEP. Using LinSca-1 c-Kit+CD34+ CD16lo panel as common myeloid progenitor population markers and Lin Sca-1 c-Kit+ CD16hi as granulocyte/monocyte progenitor markers, granulocyte/macrophage progenitor were increased more than 2-fold in bone marrow of ΔNC16A mice than WT. (C) Statistical analysis showed significantly increased granulocyte/macrophage progenitor in bone marrow of ΔNC16A mice than WT; in contrast, MEP were decreased in ΔNC16A mice than WT (n = 6). (D) Bone marrow cells were cultured by methylcellulose-based media, colonies were counted after 12 d culture. ΔNC16A mice have more granulocyte/monocyte CFUs (GM-CFU) than WT mice (n = 6). ***p < 0.001.

FIGURE 2.

Increased granulocyte/monocyte progenitor in ΔNC16A mice. (A) Diagram showing different stages of granulopoiesis and surface markers expressed in different developmental stages. HSC, hematopoietic stem cell; MEP, megakaryocyte/erythrocyte progenitor; MMP, multipotent progenitor. (B) Flow cytometry. Bone marrow cells were gated on Lin and IL-7R double-negative cells, and c-Kit–positive and Sca-1–negative cell population was used for further analysis. CD34 and CD16 were stained to distinguish granulocyte/macrophage progenitor, common myeloid progenitor, and MEP. Using LinSca-1 c-Kit+CD34+ CD16lo panel as common myeloid progenitor population markers and Lin Sca-1 c-Kit+ CD16hi as granulocyte/monocyte progenitor markers, granulocyte/macrophage progenitor were increased more than 2-fold in bone marrow of ΔNC16A mice than WT. (C) Statistical analysis showed significantly increased granulocyte/macrophage progenitor in bone marrow of ΔNC16A mice than WT; in contrast, MEP were decreased in ΔNC16A mice than WT (n = 6). (D) Bone marrow cells were cultured by methylcellulose-based media, colonies were counted after 12 d culture. ΔNC16A mice have more granulocyte/monocyte CFUs (GM-CFU) than WT mice (n = 6). ***p < 0.001.

Close modal

Knowing that ΔNC16A mice develop granulocyte hyperplasia, we then determined if a cell-autonomous or non–cell-autonomous mechanism causes the significantly elevated granulocytes by bone marrow chimera and bone marrow transplantation experiments. Taking advantage of congenic marker, bone marrows of CD45.1 WT mice and CD45.2 ΔNC16A mice were mixed at 1:1 ratio and transferred to sublethally irradiated WT CD45.1.2 recipients (Fig. 3A). After 8-wk reconstitution, there was no difference between granulocyte populations derived from WT and ΔNC16A mice in bone marrow and spleen of host animals (Fig. 3B). Gr-1+ cell populations in bone marrow, spleen, and lymph node were also comparable (Fig. 3C). By bone marrow transplantation experiments, WT mice transplanted with WT or ΔNC16A bone marrow showed normal granulopoiesis, whereas ΔNC16A mice transplanted with WT or ΔNC16A bone marrow developed granulocyte hyperplasia similar to ΔNC16A mice (Fig. 3D). These data suggested that altered granulocyte development in ΔNC16A mice is not caused by the defect in bone marrow cells themselves, but rather is caused by the dysfunction of BP180 in the radiation-resistant compartment, the MSC.

FIGURE 3.

Granulocyte hyperplasia in ΔNC16A mice were caused by extrinsic factors. (A) Diagram of bone marrow chimera experiment. Bone marrows from WT CD45.1 and ΔNC16A CD45.2 mice were mixed by 1:1 ratio and were transferred to sublethally irradiated (700 cGy) recipient CD45.1.2 mice by i.v. Eight weeks postreconstitution, surface markers were used for determining different populations of cells by flow cytometry. Granulocyte populations from WT and ΔNC16A mice in bone marrow and spleen were comparable tested by forward scatter (cell size) and side scatter (granularity) (B) and by using granulocytes specific marker Gr-1+ (C). (D) Reciprocal transplantation experiment. WT CD45.1 bone marrow was transferred to sublethally irradiated (700 cGy) recipient ΔNC16A CD45.2 mice, whereas ΔNC16A CD45.2 bone marrow was transferred to WT CD45.1 sublethally irradiated (700 cGy) recipient mice. Total blood granulocytes were counted by hematological examination 8 wk after bone marrow transplantation (50 μl of blood was used for counting). WT recipients receiving bone marrow cells either from WT or ΔNC16A mice had normal granulocyte numbers, whereas ΔNC16A recipients receiving bone marrow cells either from WT or ΔNC16A mice exhibited significantly higher number of granulocytes (n = 6). (E) Immunoblotting showed full-length BP180 in WT bone marrow and NC16A truncated BP180 in ΔNC16A bone marrow. (F) Indirect immunofluorescence exhibited anti-NC16A Ab staining in WT, and not ΔNC16A, bone marrow. Original magnification ×200. (G) Immunoblotting identified BP180 in MSC, and not granulocytes (GC) and lymphocytes (LC). ***p < 0.001.

FIGURE 3.

Granulocyte hyperplasia in ΔNC16A mice were caused by extrinsic factors. (A) Diagram of bone marrow chimera experiment. Bone marrows from WT CD45.1 and ΔNC16A CD45.2 mice were mixed by 1:1 ratio and were transferred to sublethally irradiated (700 cGy) recipient CD45.1.2 mice by i.v. Eight weeks postreconstitution, surface markers were used for determining different populations of cells by flow cytometry. Granulocyte populations from WT and ΔNC16A mice in bone marrow and spleen were comparable tested by forward scatter (cell size) and side scatter (granularity) (B) and by using granulocytes specific marker Gr-1+ (C). (D) Reciprocal transplantation experiment. WT CD45.1 bone marrow was transferred to sublethally irradiated (700 cGy) recipient ΔNC16A CD45.2 mice, whereas ΔNC16A CD45.2 bone marrow was transferred to WT CD45.1 sublethally irradiated (700 cGy) recipient mice. Total blood granulocytes were counted by hematological examination 8 wk after bone marrow transplantation (50 μl of blood was used for counting). WT recipients receiving bone marrow cells either from WT or ΔNC16A mice had normal granulocyte numbers, whereas ΔNC16A recipients receiving bone marrow cells either from WT or ΔNC16A mice exhibited significantly higher number of granulocytes (n = 6). (E) Immunoblotting showed full-length BP180 in WT bone marrow and NC16A truncated BP180 in ΔNC16A bone marrow. (F) Indirect immunofluorescence exhibited anti-NC16A Ab staining in WT, and not ΔNC16A, bone marrow. Original magnification ×200. (G) Immunoblotting identified BP180 in MSC, and not granulocytes (GC) and lymphocytes (LC). ***p < 0.001.

Close modal

To sustain this possibility, we determined whether BP180 is expressed in bone marrow and/or MSC. Using immunoblotting, anti-NC16A Ab detected the full-length BP180 in whole bone marrow protein extract of WT and not ΔNC16A mice, whereas anti–NC1-3 Ab recognized both full-length and NC16A truncated BP180 in bone marrow of both WT and ΔNC16A mice (Fig. 3E). By indirect immunofluorescence, anti-NC16A Ab stained bone marrow cryosection from WT, but not from ΔNC16A mice (Fig. 3F). More importantly, BP180 expression was only detected in cultured bone marrow–derived MSC but not in purified granulocytes and lymphocytes (Fig. 3G). Taken together, these results showed that altered granulopoiesis ΔNC16A mice is due to BP180 functional deficiency in bone marrow stromal cells instead of a cell-autonomous mechanism.

Because G-CSF is a key regulator for granulopoiesis of non–cell-autonomous mechanism and myeloid cell proliferation (25, 45), we used ELISA to quantify G-CSF expression levels both in bone marrow supernatant and serum from ΔNC16A and WT mice. ΔNC16A mice had significantly higher G-CSF levels in bone marrow and serum compared with WT mice (Fig. 4A, 4B). As a control, levels of GM-CSF in bone marrow and serum in ΔNC16A and WT mice were comparable (Fig. 4C, 4D). To determine whether activation of NF-κB pathway was also increased in ΔNC16A mice as a potential molecular mechanism underlying the increased G-CSF expression and altered granulopoiesis, we examined whether NF-κB signaling is upregulated in the bone marrow of ΔNC16A mice. Immunoblotting results showed increased level of phosphorylated IκBα in the total bone marrow cells of ΔNC16A mice as compared with WT (Fig. 4E, 4F).

FIGURE 4.

ΔNC16A mice showed elevated G-CSF level and increased activation of NF-κB signaling pathway in bone marrow. ELISA assays revealed significantly elevated levels of G-CSF in bone marrow (A) and blood (B) of ΔNC16A mice as compared with WT control, whereas levels of GM-CSF in bone marrow (C) and blood (D) of both WT and ΔNC16A mice were compatible (n = 8). (E) Protein extracts made from total bone marrow cells (whole bone marrow flush) of WT and ΔNC16A mice were analyzed by immunoblotting. Higher levels of phospho-IκBα were seen in ΔNC16A mice than in WT mice. (F) Phosphorylated IκBα and JNK were significantly increased in bone marrow of ΔNC16A mice than in WT mice as determined by densitometry analysis of the phosphorylated versus total signaling proteins (n = 4). (G) Bone marrow–derived MSC from ΔNC16A mice showed increased activation of NF-κB pathway. (H) ELISA of MSC culture medium revealed significantly higher levels of G-CSF in bone marrow–derived MSC of ΔNC16A mice as compared with WT control (n = 6). (I) Levels of bone marrow–derived MSC-released GM-CSF were similar between WT and ΔNC16A mice. *p < 0.05, **p < 0.01, ***p < 0.001.

FIGURE 4.

ΔNC16A mice showed elevated G-CSF level and increased activation of NF-κB signaling pathway in bone marrow. ELISA assays revealed significantly elevated levels of G-CSF in bone marrow (A) and blood (B) of ΔNC16A mice as compared with WT control, whereas levels of GM-CSF in bone marrow (C) and blood (D) of both WT and ΔNC16A mice were compatible (n = 8). (E) Protein extracts made from total bone marrow cells (whole bone marrow flush) of WT and ΔNC16A mice were analyzed by immunoblotting. Higher levels of phospho-IκBα were seen in ΔNC16A mice than in WT mice. (F) Phosphorylated IκBα and JNK were significantly increased in bone marrow of ΔNC16A mice than in WT mice as determined by densitometry analysis of the phosphorylated versus total signaling proteins (n = 4). (G) Bone marrow–derived MSC from ΔNC16A mice showed increased activation of NF-κB pathway. (H) ELISA of MSC culture medium revealed significantly higher levels of G-CSF in bone marrow–derived MSC of ΔNC16A mice as compared with WT control (n = 6). (I) Levels of bone marrow–derived MSC-released GM-CSF were similar between WT and ΔNC16A mice. *p < 0.05, **p < 0.01, ***p < 0.001.

Close modal

To further confirm these findings, cultured bone marrow MSCs were assayed for activation of the NF-κB signaling pathway and release of G-CSF. Cultured bone marrow MSC from ΔNC16A mice showed significant increase of NF-κB signaling pathway (Fig. 4G) and released significantly higher level of G-CSF than WT mice (Fig. 4H). In contrast, there is no significant change in GM-CSF release between ΔNC16A and WT MSCs (Fig. 4I). These results suggest that increased activation of NF-κB signaling pathway and release of G-CSF in bone marrow MSC contribute to the altered granulopoiesis and, subsequently, to elevated granulocytes in bone marrow, spleen, and peripheral blood in ΔNC16A mice.

ΔNC16A mice start showing skin lesions at 8 wk old. To rule out the possibility that altered granulopoiesis in ΔNC16A mice is secondary to skin lesion and inflammation, we crossed loxP-floxed NC16A mice with K14Cre mice to generate mice with basal keratinocyte-specific deletion of NC16A (termed K14Cre/ΔNC16A) (46). As expected, K14Cre/ΔNC16A mice showed NC16A truncated BP180 in the skin and full-length BP180 in bone marrow as evidenced by immunoblotting (Fig. 5A). Similar to ΔNC16A mice, K14Cre/ΔNC16A mice developed skin lesion starting at 8 wk of age, including skin erosion and infiltration of CD3-positive T cell and Gr-1–positive granulocytes (Supplemental Fig. 3) (46). By hematological examination, granulocyte numbers in peripheral blood of K14Cre/ΔNC16A mice were similar to WT mice, but significantly lower than ΔNC16A mice (Fig. 5B). By flow cytometry the granulocyte populations in bone marrow, spleen, and blood in K14Cre/ΔNC16A mice were comparable to WT mice (Fig. 5C). Furthermore, levels of G-CSF and GM-CSF in the serum of K14Cre/ΔNC16A mice were similar to WT mice, but G-CSF level of K14Cre/ΔNC16A mice were significantly lower than ΔNC16A mice (Fig. 5D, 5E). These results indicate that loss of BP180 function in basal keratinocytes is not the cause of the altered granulopoiesis, further supporting that altered granulopoiesis in ΔNC16A mice is due to lack of BP180 function in bone marrow MSCs.

FIGURE 5.

K14Cre/ΔNC16A mice had normal granulopoiesis. (A) Immunoblotting confirmed that keratinocyte-specific ΔNC16A (K14Cre/ΔNC16A) mice express NC16A truncated BP180 in the skin and full-length BP180 in bone marrow. (B) Hematological examination showed that K14Cre/ΔNC16A mice had significantly lower periphery granulocytes compared with ΔNC16A mice (n = 6), which is similar to WT. (C) By flow cytometry for Gr.1 and CD11b double-positive cell, K14Cre/ΔNC16A and WT mice had compatible granulocyte populations in bone marrow, spleen, and blood, which are significantly lower than Gr1+CD11b+ cells in ΔNC16A mice (n = 6). ELISA assays also revealed similar levels of serum G-CSF (D) and GM-CSF (E) between K14Cre/ΔNC16A and WT mice (n = 6, K14Cre/ΔNC16A versus ΔNC16A). ***p < 0.001.

FIGURE 5.

K14Cre/ΔNC16A mice had normal granulopoiesis. (A) Immunoblotting confirmed that keratinocyte-specific ΔNC16A (K14Cre/ΔNC16A) mice express NC16A truncated BP180 in the skin and full-length BP180 in bone marrow. (B) Hematological examination showed that K14Cre/ΔNC16A mice had significantly lower periphery granulocytes compared with ΔNC16A mice (n = 6), which is similar to WT. (C) By flow cytometry for Gr.1 and CD11b double-positive cell, K14Cre/ΔNC16A and WT mice had compatible granulocyte populations in bone marrow, spleen, and blood, which are significantly lower than Gr1+CD11b+ cells in ΔNC16A mice (n = 6). ELISA assays also revealed similar levels of serum G-CSF (D) and GM-CSF (E) between K14Cre/ΔNC16A and WT mice (n = 6, K14Cre/ΔNC16A versus ΔNC16A). ***p < 0.001.

Close modal

To further confirm that altered granulopoiesis in ΔNC16A mice originates from bone marrow stromal cells, not skin lesion, we examined 4-wk-old ΔNC16A mice when clinical skin lesions are absent. Gross examination showed that 4-wk-old ΔNC16A mice do not exhibit skin abnormalities (Supplemental Fig. 4A). Histological examination showed no skin inflammation including a lack of CD3-positive T cell and granulocyte infiltration in 4-wk-old ΔNC16A mice (Supplemental Fig. 4A). Similar to 8-wk-old ΔNC16A mice, however, 4-wk-old ΔNC16A mice also had significantly increased granulocyte numbers in peripheral blood in comparison with WT mice, as determined by hematological examination (Supplemental Fig. 4B). In addition, 4-wk-old ΔNC16A mice had more granulocytes in bone marrow and blood but not in the spleen or blood when compared with WT mice, as determined by flow cytometry using forward scatter, side scatter, Gr-1, and CD11b markers (Supplemental Fig. 4C). Taken together, these data further suggest that altered granulopoiesis in ΔNC16A mice takes place in bone marrow stromal cells because of loss of BP180 function.

If increased G-CSF is critical for altered granulopoiesis in ΔNC16A mice, then blocking G-CSF would restore normal granulocyte level in ΔNC16A mice. To test this hypothesis, anti–G-CSF neutralizing Ab or matched control Ab was administrated i.p. into ΔNC16A mice. After 12 d, the granulocyte populations in circulation and bone marrow of the injected mice were analyzed by hematological test and flow cytometry. Granulocyte number in ΔNC16A mice treated with G-CSF neutralizing Ab and not matched control Ab were restored to normal levels as shown by hematological test (Fig. 6A). Restored granulocyte population frequency was also observed in ΔNC16A mice treated with G-CSF neutralizing Ab and not matched control Ab (Fig. 6B). To further investigate if the granulocyte/macrophage progenitors were restored to normal level, LinIL-7RSCA-1c-kit+CD34+CD16+ surface marker combination panel were used by flow cytometry. As expected, the granulocyte/macrophage progenitors in anti–G-CSF neutralizing Ab-treated ΔNC16A mice were also significantly reduced and close to levels found in WT mice (Fig. 6C). These results suggest that G-CSF elevation in bone marrow is a key in myeloid hyperplasia in ΔNC16A mice.

FIGURE 6.

Blocking G-CSF and NF-κB restored normal granulopoiesis in ΔNC16A mice. WT and ΔNC16A mice were i.p. treated with anti–G-CSF neutralizing Ab or matched control Ab. PBS-treated WT and ΔNC16A mice were also included as control. Granulocytes and granulocyte/monocyte progenitor were analyzed by hematological examination and flow cytometry at day 12 posttreatment. (A) Complete cell count. Hematological test showed that anti–G-CSF Ab and not control Ab treatment reduced granulocyte population in ΔNC16A mice down to level similar to WT mice (n = 6). (B) Flow cytometry using CD11b and Gr.1 surface markers showed that anti–G-CSF neutralizing Ab and not control Ab treatment restored granulocyte level in ΔNC16A mice comparable to WT control mice. (C) Flow cytometry using Lin, Sca-1, c-Kit CD16, and CD34 markers showed that granulocyte/macrophage progenitor level in bone marrow of ΔNC16A mice was also restored to the level of WT control 12 d after anti–G-CSF Ab and not control Ab treatment. (D) WT and ∆NC16A mice were i.p. injected daily for 7 d with PDTC or PBS vehicle control. (A) Western blotting showed that PDTC significantly reduced NF-κB activation (p65 phosphorylation) in bone marrow cells of ∆NC16A mice. (E) ELISA showed that PDTC-treated ∆NC16A mice had serum G-CSF similar to WT control. (F) Hematological test showed that granulocyte level of PDTC-treated ∆NC16A mice reduced to a similar level as WT mice (n = 6). *p < 0.05, ***p < 0.001.

FIGURE 6.

Blocking G-CSF and NF-κB restored normal granulopoiesis in ΔNC16A mice. WT and ΔNC16A mice were i.p. treated with anti–G-CSF neutralizing Ab or matched control Ab. PBS-treated WT and ΔNC16A mice were also included as control. Granulocytes and granulocyte/monocyte progenitor were analyzed by hematological examination and flow cytometry at day 12 posttreatment. (A) Complete cell count. Hematological test showed that anti–G-CSF Ab and not control Ab treatment reduced granulocyte population in ΔNC16A mice down to level similar to WT mice (n = 6). (B) Flow cytometry using CD11b and Gr.1 surface markers showed that anti–G-CSF neutralizing Ab and not control Ab treatment restored granulocyte level in ΔNC16A mice comparable to WT control mice. (C) Flow cytometry using Lin, Sca-1, c-Kit CD16, and CD34 markers showed that granulocyte/macrophage progenitor level in bone marrow of ΔNC16A mice was also restored to the level of WT control 12 d after anti–G-CSF Ab and not control Ab treatment. (D) WT and ∆NC16A mice were i.p. injected daily for 7 d with PDTC or PBS vehicle control. (A) Western blotting showed that PDTC significantly reduced NF-κB activation (p65 phosphorylation) in bone marrow cells of ∆NC16A mice. (E) ELISA showed that PDTC-treated ∆NC16A mice had serum G-CSF similar to WT control. (F) Hematological test showed that granulocyte level of PDTC-treated ∆NC16A mice reduced to a similar level as WT mice (n = 6). *p < 0.05, ***p < 0.001.

Close modal

G-CSF expression is regulated by NF-κB pathway. Therefore, we tested whether blockading of NF-κB pathway also restores normal level of G-CSF and the granulopoiesis in ΔNC16A mice. ΔNC16A mice were treated by daily i.p. injection of the NF-κB inhibitor pyrolidine dithocarbamate (PDTC; 200 mg/kg in PBS) or PBS control for 7 d (54, 55). After 7 d PDTC treatment, activation of NF-κB signaling pathway as evidenced by phosphorylated NF-κB p65 was significantly reduced in bone marrow of PDTC-treated ΔNC16A mice as compared with PBS-treated ΔNC16A mice (Fig. 6D). G-CSF level (Fig. 6E) and granulocyte numbers in circulation (Fig. 6F) of PDTC-treated, but not control-treated, ΔNC16A mice were restored to normal level as WT mice. Taken together, these results suggest that NF-κB signaling pathway activation and the subsequently increased G-CSF and granulocytes are the underlying mechanism for altered granulopoiesis in ΔNC16A mice.

BP180 is well documented as a key cell–cell matrix adhesion molecule in the skin; however, whether it has other biological functions is largely unknown (9). In this study, we generated a BP180 loss-of-function mouse model (termed ΔNC16A mice) by deletion of NC16A domain of humanized NC16A mice (13). Using this model, we demonstrated that loss of BP180 function in bone marrow causes myeloid progenitor lineage skewed toward granulocyte/macrophage progenitors, leading to altered granulopoiesis with significantly increased Gr-1–positive granulocytes in immune organs including bone marrow, spleen, blood, and skin. These findings suggest that besides cell–cell matrix adhesion BP180 also plays a role in granulopoiesis.

Abnormal granulopoiesis could arise from cell-autonomous or non–cell-autonomous mechanism. Bone marrow chimera and bone marrow transplantation experiments demonstrated that the granulocyte development in ΔNC16A mice is altered in a non–cell-autonomous manner. This conclusion is further confirmed by Western blotting, showing that bone marrow stromal cells, and not granulocytes, in WT and ΔNC16A mice express full-length and truncated BP180, respectively. G-CSF is the most important regulator that drives hematopoiesis stem cells to differentiate into common myeloid progenitors and granulocyte/macrophage progenitors (30). Granulocytes express G-CSFR (29). Mice lacking G-CSF or G-CSFR show granulopenia (30, 5659). We found that increased granulocytes in ΔNC16A mice were accompanied with a significantly increased level of G-CSF in bone marrow. Bone marrow MSCs from ΔNC16A mice also secreted significantly more G-CSF than WT control mice. Furthermore, treatment with an anti–G-CSF neutralizing Ab restored normal granulopoiesis and normal granulocyte numbers in circulation in ΔNC16A mice. Taken together, these findings suggest that lack of BP180 function in MSCs leads to upregulation of G-CSF and subsequently altered granulopoiesis.

Transcription factor NF-κB is critical in regulation of G-CSF and other proinflammatory cytokines (43). Mice lacking IκBα have increased granulocyte numbers (44). Cultured human keratinocytes lacking BP180 were reported to secrete inflammatory cytokines associated with activated NF-κB signaling pathway (60). To determine whether activation of the NF-κB pathway is the underlying mechanism in BP180 functional deficiency-caused increased G-CSF and granulocytes, we analyzed the status of NF-κB signaling pathway activation in bone marrow and bone marrow MSCs of ΔNC16A mice by Western blot. We found an increased activation of NF-κB signaling pathway in ΔNC16A mice as compared with WT control mice. These results suggest that the increased activation of NF-κB signaling pathway and the release of G-CSF in bone marrow MSC contribute to the altered granulopoiesis and subsequently elevated granulocytes in bone marrow, spleen, and peripheral blood. JNK cross-talks with NF-κB pathway and plays an important role in cell survival (61). Phosphorylated JNK is also significantly upregulated in ΔNC16A mice (Fig. 4F), suggesting potential involvement of JNK in altered granulopoiesis in ΔNC16A mice. Future studies are needed to elucidate the role of JNK and functional interaction with NF-κB signaling pathway in the granulopoiesis in ΔNC16A mice.

Granulopoiesis could result from intrinsic gene regulation or extrinsic environment cues (21, 62, 63). Periphery infection could lead to unbalanced homeostasis of hematopoietic stem cells, causing emergency granulopoiesis (34). ΔNC16A mice develop skin lesions starting at the age of 8 wk old. This includes reduced skin barrier function that makes ΔNC16A mice with skin lesions more prone to infection and development of spontaneous skin inflammation with pruritus; evidence shows infiltration of inflammatory immune cells and an increase of inflammatory cytokines, such as thymic stromal lymphopoietin (TSLP) (46). To rule out the possibility that altered granulopoiesis in 8-wk-old or older ΔNC16A mice is caused by or partially contributed to by skin lesions/skin inflammation, we determined whether basal keratinocyte-specific NC16A deletion (termed K14Cre/ΔNC16A) also leads to increased granulocytes in immune organs as seen in whole body ΔNC16A mice. We found that K14Cre/ΔNC16A mice phenocopied skin lesion of ΔNC16A mice but had normal granulopoiesis and compatible granulocyte numbers in bone marrow, spleen, and blood as WT control mice. To further sustain that skin lesion/skin inflammation is not the main cause for the altered granulopoiesis in ΔNC16A mice, we quantified granulocytes in 4-wk-old ΔNC16A mice when no skin abnormality presents clinically or histologically. Like the 8-wk-old ΔNC16A mice, 4-wk-old ΔNC16A mice also exhibited significantly increased granulocytes as compared with WT control. In addition, the NF-κB signaling pathway is also related to TSLP expression by NF-κB binding sites present on the TSLP promoter (64). Furthermore, NF-κB signaling pathway activation leads to secretion of inflammatory cytokines such as TNF-α, Il-1β, and IL-13, which results in increased TSLP expression (65). However, this mechanism needs further investigation in ΔNC16A mice. These results suggest that BP180 in bone marrow regulates normal granulopoiesis.

BP180 function loss or dysfunction can be caused by anti-BP180 autoantibodies in BP or BP180 gene mutations in JEB. Our ΔNC16A mice reflect more on JEB instead of BP. JEB is a rare genetic disease and present marked skin fragility, blister formation following minor trauma, along with eosinophilia, eosinophils and/or neutrophils infiltrated in the dermis in some cases (14, 17). BP180 functional–deficient mice generated by another group also exhibit a similar skin clinical phenotypes and a significantly increased granulocyte infiltration in the skin (20). However, human patients with JEB have not been reported to have neutrophilia. This discrepancy could be explained by several scenarios: 1) JEB covers a wide spectrum of symptoms among patients, especially because of the BP180 gene mutations (17). Our NC16A truncated mice still have a partially functional BP180 and may reflect a much milder form of JEB (46, 66); 2) JEB is a very rare genetic disease with limited cases reported so far (67). New JEB patients remain to be identified to have a partially functional BP180 and neutrophilia as seen in our NC16A truncated mice; and 3) human and mice may differ in term of BP180 involvement in granulopoiesis. For example, BP180 regulated granulopoiesis in mice might require other modifier(s) that lack in human. Therefore, our present work provides reverse-genetic evidence linking BP180 function and granulopoiesis. Future clinic studies may support our current findings.

In summary, our results demonstrate that BP180 as a cell–cell matrix adhesion molecule regulates granulopoiesis. BP180 in bone marrow stromal cells acts on the NF-κB signaling pathway. Loss of BP180 function in these cells leads to increase NF-κB pathway activation and subsequent upregulation of G-CSF and uncontrolled granulopoiesis. Previous reports showed that cultured human and mouse keratinocytes lacking BP180 function secrete inflammatory cytokines associated with activated NF-κB signaling pathway (60, 66). It is of great interest and significance to know if BP180 loss-of-function human marrow MSC behave the same as NC16A-fecifient mouse MSC and if BP180 regulates NF-κB pathway in other cell lineages besides MSC and keratinocytes. Our future studies will directly address these issues by using gene editing to determine the role of BP180 in the signaling pathways of human cells. Our findings point to new directions for increased understanding of the pathophysiology of diseases associated with BP180 loss of function and abnormal granulopoiesis. However, it remains unclear whether BP180 plays a different role at different stages of granulocyte development and how BP180 regulates NF-κB pathway at the molecular level.

We thank Dr. Dennis Roop at the University of Colorado at Denver for providing K14Cre mice.

This work was supported by the National Institutes of Health (R01 AI40768 and R01 AI61430 to Z.L. and P01CA206980 to N.E.T), and a Graduate School, University of North Carolina at Chapel Hill Cancer Center Research Award (to Z.L.) and Graduate School Dissertation Completion Fellowship (to L.L.).

The online version of this article contains supplemental material.

Abbreviations used in this article:

BP

bullous pemphigoid

JEB

junctional epidermolysis bullosa

MSC

mesenchymal stem cell

NC

noncollagen

PDTC

pyrolidine dithocarbamate

TSLP

thymic stromal lymphopoietin

WT

wild-type.

1
Hammers
,
C. M.
,
J. R.
Stanley
.
2016
.
Mechanisms of disease: pemphigus and bullous pemphigoid.
Annu. Rev. Pathol.
11
:
175
197
.
2
Koster
,
J.
,
D.
Geerts
,
B.
Favre
,
L.
Borradori
,
A.
Sonnenberg
.
2003
.
Analysis of the interactions between BP180, BP230, plectin and the integrin alpha6beta4 important for hemidesmosome assembly.
J. Cell Sci.
116
:
387
399
.
3
Hopkinson
,
S. B.
,
J. C.
Jones
.
2000
.
The N terminus of the transmembrane protein BP180 interacts with the N-terminal domain of BP230, thereby mediating keratin cytoskeleton anchorage to the cell surface at the site of the hemidesmosome.
Mol. Biol. Cell
11
:
277
286
.
4
Margadant
,
C.
,
E.
Frijns
,
K.
Wilhelmsen
,
A.
Sonnenberg
.
2008
.
Regulation of hemidesmosome disassembly by growth factor receptors.
Curr. Opin. Cell Biol.
20
:
589
596
.
5
Diaz
,
L. A.
,
H.
Ratrie
III
,
W. S.
Saunders
,
S.
Futamura
,
H. L.
Squiquera
,
G. J.
Anhalt
,
G. J.
Giudice
.
1990
.
Isolation of a human epidermal cDNA corresponding to the 180-kD autoantigen recognized by bullous pemphigoid and herpes gestationis sera. Immunolocalization of this protein to the hemidesmosome.
J. Clin. Invest.
86
:
1088
1094
.
6
Liu
,
Z.
,
L. A.
Diaz
,
J. L.
Troy
,
A. F.
Taylor
,
D. J.
Emery
,
J. A.
Fairley
,
G. J.
Giudice
.
1993
.
A passive transfer model of the organ-specific autoimmune disease, bullous pemphigoid, using antibodies generated against the hemidesmosomal antigen, BP180.
J. Clin. Invest.
92
:
2480
2488
.
7
Foldes
,
C.
,
D.
Wallach
,
E.
Aubinière
,
M. D.
Vignon-Pennamen
,
F.
Cottenot
.
1988
.
Generalized atrophic benign form of junctional epidermolysis bullosa.
Dermatologica
176
:
83
90
.
8
Darling
,
T. N.
,
J. W.
Bauer
,
H.
Hintner
,
K. B.
Yancey
.
1997
.
Generalized atrophic benign epidermolysis bullosa.
Adv. Dermatol.
13
:
87
119
,
discussion 120
.
9
Leighty
,
L.
,
N.
Li
,
L. A.
Diaz
,
Z.
Liu
.
2007
.
Experimental models for the autoimmune and inflammatory blistering disease, Bullous pemphigoid.
Arch. Dermatol. Res.
299
:
417
422
.
10
Nousari
,
H. C.
,
G. J.
Anhalt
.
1999
.
Pemphigus and bullous pemphigoid.
Lancet
354
:
667
672
.
11
Giudice
,
G. J.
,
D. J.
Emery
,
B. D.
Zelickson
,
G. J.
Anhalt
,
Z.
Liu
,
L. A.
Diaz
.
1993
.
Bullous pemphigoid and herpes gestationis autoantibodies recognize a common non-collagenous site on the BP180 ectodomain.
J. Immunol.
151
:
5742
5750
.
12
Zillikens
,
D.
,
P. A.
Rose
,
S. D.
Balding
,
Z.
Liu
,
M.
Olague-Marchan
,
L. A.
Diaz
,
G. J.
Giudice
.
1997
.
Tight clustering of extracellular BP180 epitopes recognized by bullous pemphigoid autoantibodies.
J. Invest. Dermatol.
109
:
573
579
.
13
Liu
,
Z.
,
W.
Sui
,
M.
Zhao
,
Z.
Li
,
N.
Li
,
R.
Thresher
,
G. J.
Giudice
,
J. A.
Fairley
,
C.
Sitaru
,
D.
Zillikens
, et al
.
2008
.
Subepidermal blistering induced by human autoantibodies to BP180 requires innate immune players in a humanized bullous pemphigoid mouse model.
J. Autoimmun.
31
:
331
338
.
14
Bruckner-Tuderman
,
L.
,
J. A.
McGrath
,
E. C.
Robinson
,
J.
Uitto
.
2013
.
Progress in epidermolysis bullosa research: summary of DEBRA International Research Conference 2012.
J. Invest. Dermatol.
133
:
2121
2126
.
15
Jonkman
,
M. F.
,
M. C.
de Jong
,
K.
Heeres
,
H. H.
Pas
,
J. B.
van der Meer
,
K.
Owaribe
,
A. M.
Martinez de Velasco
,
C. M.
Niessen
,
A.
Sonnenberg
.
1995
.
180-kD bullous pemphigoid antigen (BP180) is deficient in generalized atrophic benign epidermolysis bullosa.
J. Clin. Invest.
95
:
1345
1352
.
16
Yuen
,
W. Y.
,
H. H.
Pas
,
R. J.
Sinke
,
M. F.
Jonkman
.
2011
.
Junctional epidermolysis bullosa of late onset explained by mutations in COL17A1.
Br. J. Dermatol.
164
:
1280
1284
.
17
Kiritsi
,
D.
,
J. S.
Kern
,
H.
Schumann
,
J.
Kohlhase
,
C.
Has
,
L.
Bruckner-Tuderman
.
2011
.
Molecular mechanisms of phenotypic variability in junctional epidermolysis bullosa.
J. Med. Genet.
48
:
450
457
.
18
Mabuchi
,
E.
,
N.
Umegaki
,
H.
Murota
,
T.
Nakamura
,
K.
Tamai
,
I.
Katayama
.
2007
.
Oral steroid improves bullous pemphigoid-like clinical manifestations in non-Herlitz junctional epidermolysis bullosa with COL17A1 mutation.
Br. J. Dermatol.
157
:
596
598
.
19
Cifuentes
,
L.
,
D.
Kiritsi
,
W.
Chen
,
J.
Pennino
,
J.
Ring
,
S.
Weidinger
,
C.
Has
.
2013
.
A case of junctional epidermolysis bullosa with prurigo-like lesions and reduction of collagen XVII and filaggrin.
Br. J. Dermatol.
169
:
195
198
.
20
Hurskainen
,
T.
,
N.
Kokkonen
,
R.
Sormunen
,
J.
Jackow
,
S.
Löffek
,
R.
Soininen
,
C.-W.
Franzke
,
L.
Bruckner-Tuderman
,
K.
Tasanen
.
2015
.
Deletion of the major bullous pemphigoid epitope region of collagen XVII induces blistering, autoimmunization, and itching in mice.
J. Invest. Dermatol.
135
:
1303
1310
.
21
Friedman
,
A. D.
2007
.
Transcriptional control of granulocyte and monocyte development.
Oncogene
26
:
6816
6828
.
22
Eaves
,
C. J.
2015
.
Hematopoietic stem cells: concepts, definitions, and the new reality.
Blood
125
:
2605
2613
.
23
Göttgens
,
B.
2015
.
Regulatory network control of blood stem cells.
Blood
125
:
2614
2620
.
24
Boulais
,
P. E.
,
P. S.
Frenette
.
2015
.
Making sense of hematopoietic stem cell niches.
Blood
125
:
2621
2629
.
25
Amulic
,
B.
,
C.
Cazalet
,
G. L.
Hayes
,
K. D.
Metzler
,
A.
Zychlinsky
.
2012
.
Neutrophil function: from mechanisms to disease.
Annu. Rev. Immunol.
30
:
459
489
.
26
Nauseef
,
W. M.
,
N.
Borregaard
.
2014
.
Neutrophils at work.
Nat. Immunol.
15
:
602
611
.
27
Panopoulos
,
A. D.
,
S. S.
Watowich
.
2008
.
Granulocyte colony-stimulating factor: molecular mechanisms of action during steady state and ‘emergency’ hematopoiesis.
Cytokine
42
:
277
288
.
28
Yoda
,
M.
,
T.
Kimura
,
T.
Tohmonda
,
S.
Uchikawa
,
T.
Koba
,
J.
Takito
,
H.
Morioka
,
M.
Matsumoto
,
D. C.
Link
,
K.
Chiba
, et al
.
2011
.
Dual functions of cell-autonomous and non-cell-autonomous ADAM10 activity in granulopoiesis.
Blood
118
:
6939
6942
.
29
Beekman
,
R.
,
I. P.
Touw
.
2010
.
G-CSF and its receptor in myeloid malignancy.
Blood
115
:
5131
5136
.
30
Basu
,
S.
,
A.
Dunn
,
A.
Ward
.
2002
.
G-CSF: function and modes of action (Review).
Int. J. Mol. Med.
10
:
3
10
.
31
Bajrami
,
B.
,
H.
Zhu
,
H. J.
Kwak
,
S.
Mondal
,
Q.
Hou
,
G.
Geng
,
K.
Karatepe
,
Y. C.
Zhang
,
C.
Nombela-Arrieta
,
S. Y.
Park
, et al
.
2016
.
G-CSF maintains controlled neutrophil mobilization during acute inflammation by negatively regulating CXCR2 signaling.
J. Exp. Med.
213
:
1999
2018
.
32
Boettcher
,
S.
,
R. C.
Gerosa
,
R.
Radpour
,
J.
Bauer
,
F.
Ampenberger
,
M.
Heikenwalder
,
M.
Kopf
,
M. G.
Manz
.
2014
.
Endothelial cells translate pathogen signals into G-CSF-driven emergency granulopoiesis.
Blood
124
:
1393
1403
.
33
Tomicic
,
M.
,
M.
Starcevic
,
V.
Zach
,
J.
Bingulac-Popovic
,
Z.
Hundric-Haspl
.
2009
.
A case of neonatal neutropenia due to anti-Fc gamma receptor IIIb isoantibodies treated with recombinant human granulocyte colony stimulating factor.
Case Rep. Med.
2009
:
717545
.
34
Manz
,
M. G.
,
S.
Boettcher
.
2014
.
Emergency granulopoiesis.
Nat. Rev. Immunol.
14
:
302
314
.
35
Akala
,
O. O.
,
M. F.
Clarke
.
2006
.
Hematopoietic stem cell self-renewal.
Curr. Opin. Genet. Dev.
16
:
496
501
.
36
Pajcini
,
K. V.
,
N. A.
Speck
,
W. S.
Pear
.
2011
.
Notch signaling in mammalian hematopoietic stem cells.
Leukemia
25
:
1525
1532
.
37
Yao
,
D.
,
Y.
Huang
,
X.
Huang
,
W.
Wang
,
Q.
Yan
,
L.
Wei
,
W.
Xin
,
S.
Gerson
,
P.
Stanley
,
J. B.
Lowe
,
L.
Zhou
.
2011
.
Protein O-fucosyltransferase 1 (Pofut1) regulates lymphoid and myeloid homeostasis through modulation of Notch receptor ligand interactions.
Blood
117
:
5652
5662
.
38
Khalaj
,
M.
,
C. Y.
Park
.
2014
.
Notch one up to stroma: endothelial notch prevents inflammation and myeloproliferation.
Cell Stem Cell
15
:
1
2
.
39
Calvi
,
L. M.
,
D. C.
Link
.
2015
.
The hematopoietic stem cell niche in homeostasis and disease.
Blood
126
:
2443
2451
.
40
Wang
,
L.
,
H.
Zhang
,
S.
Rodriguez
,
L.
Cao
,
J.
Parish
,
C.
Mumaw
,
A.
Zollman
,
M. M.
Kamoka
,
J.
Mu
,
D. Z.
Chen
, et al
.
2014
.
Notch-dependent repression of miR-155 in the bone marrow niche regulates hematopoiesis in an NF-κB-dependent manner.
Cell Stem Cell
15
:
51
65
.
41
Zhao
,
C.
,
Y.
Xiu
,
J.
Ashton
,
L.
Xing
,
Y.
Morita
,
C. T.
Jordan
,
B. F.
Boyce
.
2012
.
Noncanonical NF-κB signaling regulates hematopoietic stem cell self-renewal and microenvironment interactions.
Stem Cells
30
:
709
718
.
42
Napetschnig
,
J.
,
H.
Wu
.
2013
.
Molecular basis of NF-κB signaling.
Annu. Rev. Biophys.
42
:
443
468
.
43
Grigoriadis
,
G.
,
Y.
Zhan
,
R. J.
Grumont
,
D.
Metcalf
,
E.
Handman
,
C.
Cheers
,
S.
Gerondakis
.
1996
.
The Rel subunit of NF-kappaB-like transcription factors is a positive and negative regulator of macrophage gene expression: distinct roles for Rel in different macrophage populations.
EMBO J.
15
:
7099
7107
.
44
Beg
,
A. A.
,
W. C.
Sha
,
R. T.
Bronson
,
D.
Baltimore
.
1995
.
Constitutive NF-kappa B activation, enhanced granulopoiesis, and neonatal lethality in I kappa B alpha-deficient mice.
Genes Dev.
9
:
2736
2746
.
45
Mankan
,
A. K.
,
O.
Canli
,
S.
Schwitalla
,
P.
Ziegler
,
J.
Tschopp
,
T.
Korn
,
F. R.
Greten
.
2011
.
TNF-alpha-dependent loss of IKKbeta-deficient myeloid progenitors triggers a cytokine loop culminating in granulocytosis.
Proc. Natl. Acad. Sci. USA
108
:
6567
6572
.
46
Zhang
,
Y.
,
B.-J.
Hwang
,
Z.
Liu
,
N.
Li
,
K.
Lough
,
S. E.
Williams
,
J.
Chen
,
S. W.
Burette
,
L. A.
Diaz
,
M. A.
Su
, et al
.
2018
.
BP180 dysfunction triggers spontaneous skin inflammation in mice.
Proc. Natl. Acad. Sci. USA
115
:
6434
6439
.
47
Spangrude
,
G. J.
2008
.
Assessment of lymphocyte development in radiation bone marrow chimeras.
Curr. Protoc. Immunol.
81
:
4.6.1
4.6.9
.
48
Maryanovich
,
M.
,
A.
Gross
.
2014
.
Competitive bone-marrow transplantations.
Bio Protoc.
4
: e1145.
49
Petit
,
I.
,
M.
Szyper-Kravitz
,
A.
Nagler
,
M.
Lahav
,
A.
Peled
,
L.
Habler
,
T.
Ponomaryov
,
R. S.
Taichman
,
F.
Arenzana-Seisdedos
,
N.
Fujii
, et al
.
2002
.
G-CSF induces stem cell mobilization by decreasing bone marrow SDF-1 and up-regulating CXCR4. [Published erratum appears in 2002 Nat. Immunol. 3: 787.]
Nat. Immunol.
3
:
687
694
.
50
Soleimani
,
M.
,
S.
Nadri
.
2009
.
A protocol for isolation and culture of mesenchymal stem cells from mouse bone marrow.
Nat. Protoc.
4
:
102
106
.
51
Kawakatsu
,
M.
,
Y.
Urata
,
S.
Goto
,
Y.
Ono
,
T. S.
Li
.
2013
.
Placental extract protects bone marrow-derived stem/progenitor cells against radiation injury through anti-inflammatory activity.
J. Radiat. Res. (Tokyo)
54
:
268
276
.
52
Wilson
,
A.
,
A.
Trumpp
.
2006
.
Bone-marrow haematopoietic-stem-cell niches.
Nat. Rev. Immunol.
6
:
93
106
.
53
Hoggatt
,
J.
,
Y.
Kfoury
,
D. T.
Scadden
.
2016
.
Hematopoietic stem cell niche in health and disease.
Annu. Rev. Pathol.
11
:
555
581
.
54
Schreck
,
R.
,
B.
Meier
,
D. N.
Männel
,
W.
Dröge
,
P. A.
Baeuerle
.
1992
.
Dithiocarbamates as potent inhibitors of nuclear factor kappa B activation in intact cells.
J. Exp. Med.
175
:
1181
1194
.
55
Liu
,
S. F.
,
X.
Ye
,
A. B.
Malik
.
1999
.
Inhibition of NF-kappaB activation by pyrrolidine dithiocarbamate prevents in vivo expression of proinflammatory genes.
Circulation
100
:
1330
1337
.
56
Lieschke
,
G. J.
,
D.
Grail
,
G.
Hodgson
,
D.
Metcalf
,
E.
Stanley
,
C.
Cheers
,
K. J.
Fowler
,
S.
Basu
,
Y. F.
Zhan
,
A. R.
Dunn
.
1994
.
Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization.
Blood
84
:
1737
1746
.
57
Liu
,
F.
,
H. Y.
Wu
,
R.
Wesselschmidt
,
T.
Kornaga
,
D. C.
Link
.
1996
.
Impaired production and increased apoptosis of neutrophils in granulocyte colony-stimulating factor receptor-deficient mice.
Immunity
5
:
491
501
.
58
Roberts
,
A. W.
2005
.
G-CSF: a key regulator of neutrophil production, but that’s not all!
Growth Factors
23
:
33
41
.
59
Basu
,
S.
,
G.
Hodgson
,
M.
Katz
,
A. R.
Dunn
.
2002
.
Evaluation of role of G-CSF in the production, survival, and release of neutrophils from bone marrow into circulation.
Blood
100
:
854
861
.
60
Van den Bergh
,
F.
,
S. L.
Eliason
,
B. T.
Burmeister
,
G. J.
Giudice
.
2012
.
Collagen XVII (BP180) modulates keratinocyte expression of the proinflammatory chemokine, IL-8.
Exp. Dermatol.
21
:
605
611
.
61
Nakano
,
H.
2004
.
Signaling crosstalk between NF-kappaB and JNK.
Trends Immunol.
25
:
402
405
.
62
Strydom
,
N.
,
S. M.
Rankin
.
2013
.
Regulation of circulating neutrophil numbers under homeostasis and in disease.
J. Innate Immun.
5
:
304
314
.
63
Christopher
,
M. J.
,
D. C.
Link
.
2007
.
Regulation of neutrophil homeostasis.
Curr. Opin. Hematol.
14
:
3
8
.
64
Cultrone
,
A.
,
T.
de Wouters
,
O.
Lakhdari
,
D.
Kelly
,
I.
Mulder
,
E.
Logan
,
N.
Lapaque
,
J.
Doré
,
H. M.
Blottière
.
2013
.
The NF-κB binding site located in the proximal region of the TSLP promoter is critical for TSLP modulation in human intestinal epithelial cells.
Eur. J. Immunol.
43
:
1053
1062
.
65
Redhu
,
N. S.
,
A.
Saleh
,
A. J.
Halayko
,
A. S.
Ali
,
A. S.
Gounni
.
2011
.
Essential role of NF-κB and AP-1 transcription factors in TNF-α-induced TSLP expression in human airway smooth muscle cells.
Am. J. Physiol. Lung Cell. Mol. Physiol.
300
:
L479
L485
.
66
Hwang
,
B. J.
,
Y.
Zhang
,
J. M.
Brozowski
,
Z.
Liu
,
S.
Burette
,
K.
Lough
,
C. C.
Smith
,
Y.
Shan
,
J.
Chen
,
N.
Li
, et al
.
2019
.
The dysfunction of BP180/collagen XVII in keratinocytes promotes melanoma progression.
Oncogene
38
:
7491
7503
.
67
Fine
,
J.-D.
2010
.
Inherited epidermolysis bullosa.
Orphanet J. Rare Dis.
5
:
12
.

The authors have no financial conflicts of interest.

Supplementary data