G-CSF drives autoinflammation in APLAID

Mulazzani, Elisabeth; Kong, Klara; Aróstegui, Juan I.; Ng, Ashley P.; Ranathunga, Nishika; Abeysekera, Waruni; Garnham, Alexandra L.; Ng, Sze-Ling; Baker, Paul J.; Jackson, Jacob T.; Lich, John D.; Hibbs, Margaret L.; Wicks, Ian P.; Louis, Cynthia; Masters, Seth L.

doi:10.1038/s41590-023-01473-6

Download PDF

Article
Open access
Published: 30 March 2023

G-CSF drives autoinflammation in APLAID

Nature Immunology volume 24, pages 814–826 (2023)Cite this article

6940 Accesses
10 Citations
23 Altmetric
Metrics details

Subjects

Abstract

Missense mutations in PLCG2 can cause autoinflammation with phospholipase C gamma 2-associated antibody deficiency and immune dysregulation (APLAID). Here, we generated a mouse model carrying an APLAID mutation (p.Ser707Tyr) and found that inflammatory infiltrates in the skin and lungs were only partially ameliorated by removing inflammasome function via the deletion of caspase-1. Also, deleting interleukin-6 or tumor necrosis factor did not fully prevent APLAID mutant mice from autoinflammation. Overall, these findings are in accordance with the poor response individuals with APLAID have to treatments that block interleukin-1, JAK1/2 or tumor necrosis factor. Cytokine analysis revealed increased granulocyte colony-stimulating factor (G-CSF) levels as the most distinct feature in mice and individuals with APLAID. Remarkably, treatment with a G-CSF antibody completely reversed established disease in APLAID mice. Furthermore, excessive myelopoiesis was normalized and lymphocyte numbers rebounded. APLAID mice were also fully rescued by bone marrow transplantation from healthy donors, associated with reduced G-CSF production, predominantly from non-hematopoietic cells. In summary, we identify APLAID as a G-CSF-driven autoinflammatory disease, for which targeted therapy is feasible.

Metabolic rewiring promotes anti-inflammatory effects of glucocorticoids

Article 10 April 2024

An autoantibody signature predictive for multiple sclerosis

Article 19 April 2024

Depleting myeloid-biased haematopoietic stem cells rejuvenates aged immunity

Article 27 March 2024

Main

APLAID is a rare, autoinflammatory syndrome characterized by both systemic inflammation and mild to severe immunodeficiency¹. Clinical and laboratory features of APLAID include recurrent blistering skin lesions, pulmonary manifestations, joint pain, inflammatory eye and bowel disease, accompanied by a reduced number of class-switched memory B cells and recurrent bacterial infections¹. Disease onset occurs typically during infancy or early childhood with varying degrees of severity. A total of ten clinical cases of APLAID have been reported (Extended Data Table 1)^1,2,3,4,5,6. To date, there is no effective therapy for APLAID and treatment with tumor necrosis factor (TNF) inhibitors and the interleukin (IL)-1 receptor antagonist anakinra only partially improved the disease course⁵.

APLAID is a dominantly inherited disease caused by monoallelic missense mutations in the PLCG2 gene¹. Aside from APLAID, other inherited PLCG2 mutations are identified in phospholipase C gamma 2 (PLCγ2)-associated antibody deficiency and immune dysregulation (PLAID)⁷, while acquired mutations and variants have been reported in cancer⁸ and neurodegenerative diseases⁹.

PLCγ2 is highly conserved among species and characterized by a multidomain insert between the X Box and Y Box, which consists of a split PH domain, N-terminal SH2 (nSH2) domain, C-terminal SH (cSH2) domain and an SH3 domain. The two interaction surfaces (the split PH/catalytic domain and the cSH2/C2 domain) keep PLCγ2 in an autoinhibited form¹⁰. Among the reported APLAID cases, a total of six different mutations have been identified, almost all located within the regulatory region, resulting in failure of autoinhibition, constitutive phospholipase activity and an increased production of both intracellular inositol-1,4,5-trisphosphate (IP₃) and calcium¹.

PLCγ2 is triggered upon activation and phosphorylation of non-receptor tyrosine kinases (such as SYK) or Tec kinases (such as BTK)¹¹, resulting in phosphorylation of PLCγ2 at multiple sites¹². In turn, PLCγ2 converts phospholipid phosphatidylinositol-4,5-bisphosphate (PIP₂) into the second messengers diacylglycerol (DAG) and IP₃¹³, resulting in the release of endoplasmic reticulum-stored calcium. The role of calcium as a second messenger is well established ranging from stimulation of cell proliferation and cell growth to lymphocyte activation¹⁴. However, it is the hematopoietic cell type that determines the consequence of PLCγ2 activation. In the context of APLAID, impaired B cell differentiation and enhanced myelopoiesis are the key immunological features (Extended Data Table 1), which can be explained by the critical role of PLCγ2 in both cell types. While B cell receptor signaling requires the cSH2 and C2 domain interfaces of PLCγ2 to associate with BLNK and the B cell signalosome¹⁵, thus affecting survival of mature B cells and antibody production, myeloid cells also depend on PLCγ2 for myeloid differentiation and hematopoietic development^16,17.

The potential mechanism by which autoinflammation is promoted in APLAID remains elusive. In vitro studies have implicated the NLRP3 inflammasome as patients’ peripheral blood mononuclear cells secreted increased levels of IL-1β in response to lipopolysaccharide priming alone, and this effect was attenuated by using a PLC inhibitor, intracellular calcium blockers or an adenylate cyclase activator¹⁸. Others found that PLCγ2 variants activate the NLRP3 inflammasome through the noncanonical pathway that requires ATP or nigericin as stimuli after lipopolysaccharide priming⁵. However, patients do not have a robust and durable response to blocking the inflammasome cytokine IL-1β⁵.

N-ethyl-N-nitrosourea (ENU) mutagenesis studies have created mice with spontaneous mutations in Plcg2. The initially described model was the Ali5 mouse, which carried a heterozygous gain-of-function mutation in Plcg2 at p.Asp993Gly¹⁹, leading to severe spontaneous inflammation and autoimmunity. The second mouse model, Ali14, bore a heterozygous gain-of-function mutation at p.Tyr495Cys of the split PH domain of Plcg2, and exhibited spontaneous hind paw swelling/inflammation, hypergammaglobulinemia and infertility²⁰. Although these models improved our mechanistic understanding, further in vivo studies were hampered because the phenotype did not transfer to the C57BL/6 background.

To dissect the pathogenicity of APLAID disease in an in vivo setting and to identify therapeutic regimes, we generated a de novo mouse model, which carries the exact human p.Ser707Tyr mutation as found in APLAID.

Results

Phenotype of mice encoding an APLAID mutation in Plcy2 (p.Ser707Tyr)

We generated C57BL/6 mice in which the APLAID mutation p.Ser707Tyr was introduced into the Plcg2 locus via targeting methodology, after a Neomycin cassette flanked by Frt sites that prevents expression of the mutant allele until Flp was introduced into the system (Fig. 1a). Heterozygous Plcg2 offspring were then bred with C57BL/6 Flp deleter mice, which allowed the excision of the Neomycin stop selection cassette to obtain Plcg2^S707Y/+ mice. Phenotypically, Plcg2^S707Y/+ mice display dry, flaky skin at birth. As disease progresses, skin inflammation develops on the paws, ears and tail (Fig. 1b). Post-weaning severity of skin lesions was assessed by a clinical score ranging from 0 (no clinical signs) to 5 (clinical endpoint; Fig. 1c and Methods). Plcg2^S707Y/+ pups exhibited lower body weight over time compared to littermate controls (Fig. 1d) and developed splenomegaly (Fig. 1e). Computed tomography (CT) scans revealed reduced bone mass compared to littermate controls (Extended Data Fig. 1a). Lifespan was dramatically shortened in Plcg2^S707Y/+ mice, which typically died within 6 weeks of age (Fig. 1f). Histologic examination of skin, lung and gut tissues revealed infiltrations of leukocytes (CD45⁺; Extended Data Fig. 1b). While T (CD3⁺Ki67) and B (B220⁺) cells remained unaltered (Extended Data Fig. 1b) in comparison to wild-type (WT) littermate controls (Extended Data Fig. 1c), neutrophils (MPO⁺) and macrophages (F4/80⁺) were found to be the dominant immune cell type (Fig. 1g) in the different analyzed organs of the APLAID disease model. Notably, neutrophils (MPO⁺) resided rather in the tissue than in the blood vessel endothelium (Extended Data Fig. 1d). Automated peripheral blood cell analysis (ADVIA) showed a slight increase of neutrophils and a mild decrease in lymphocytes during disease onset compared to controls (Fig. 1h). There was no difference in platelet and red blood cell counts between Plcg2^S707/+ mice and controls (Extended Data Fig. 1e–g). Fluorescence-activated cell sorting (FACS) analysis of the skin revealed an expansion of monocytes/macrophages and increased neutrophil counts in the lung during disease onset compared to Plcg2^+/+ mice, while neutrophil counts were decreased in the bone marrow (BM; Fig. 1i,j). At peak of disease (approximately 6 weeks of age), skin and lung of APLAID mice showed a notable neutrophil infiltration compared to Plcg2^+/+ mice, but no difference of neutrophil or myeloid cell counts in blood and BM was observed (Fig. 1k–m). We identified comparable T cell numbers in thymus and spleen and slightly reduced B cell counts in the BM and spleens of Plcg2^S707Y/+ mice (Fig. 1m–o). Unlike participants, APLAID mice showed unaltered or reduced (IgG2a) IgG levels in plasma (Fig. 1p).

**Fig. 1: Phenotype of mice encoding an APLAID mutation in *Plcg2* (p.Ser707Tyr).**

Collectively, these data indicate that the clinical phenotype of Plcg2^S707Y/+ mice recapitulates much of the human disease APLAID, which is characterized by neutrophilic skin and lung inflammation together with mild immunodeficiency. However, in contrast to individuals with APLAID, Plcg2^S707/+ mice lack inflammatory eye diseases and a pronounced immunoglobulin reduction in the plasma.

Minimal rescue of APLAID mice after deletion of IL-6, caspase-1 or TNF

Based on previous in vitro results, we examined whether autoinflammation in APLAID is driven by IL-6, inflammasome or TNF. Surprisingly, Plcg2^S707Y/+IL-6^−/−, Plcg2^S707Y/+caspase-1^−/− and Plcg2^S707Y/+TNF^−/− mice showed persistent skin inflammation as determined by the severity of the skin score (Fig. 2a). Plcg2^S707Y/+IL-6^−/− mice had an increased weight gain compared to Plcg2^S707Y/+ mice (Fig. 2b). Post-weaning Plcg2^S707Y/+caspase-1^−/− appeared runted and displayed a stunted weight gain compared to Plcg2^S707Y/+ mice (Fig. 2b). Plcg2^S707Y/+TNF^−/− exhibited no difference with regards to body weight compared to Plcg2^S707Y/+ mice (Fig. 2b). Although Plcg2^S707Y/+TNF^−/− mutants outlived Plcg2^S707Y/+ mice by about 12 d (Fig. 2c), Plcg2^S707Y/+IL-6^−/−, Plcg2^S707Y/+TNF^−/− and Plcg2^S707Y/+caspase-1^−/− mice all had decreased life expectancy compared to healthy littermate controls (Fig. 2c). Histologic examination revealed infiltration of leukocytes (CD45⁺), neutrophils (MPO⁺) and macrophages (F4/80⁺) into skin and lung (Fig. 2d), but not T and B cells (B220⁺; Extended Data Fig. 2a). ADVIA analysis showed a persistent neutrophilia in Plcg2^S707Y/+TNF^−/− mice and a mild decrease in lymphocytes among all mutants and intercrossed mice (Fig. 2e). No major differences in platelets or red blood cell numbers were observed (Extended Data Fig. 2b–d). Abnormal IgG2a subtype levels were present as determined by ELISA (Fig. 2f). All clinical features of different knockout strains are summarized in Extended Data Table 2. Interestingly, a multiplex assay in the plasma of Plcg2 mice and Plcg2 mutant/cytokine knockout mice revealed a marked increase in G-CSF levels and, to a lesser extent, the cytokines IL-12p40 and eotaxin (Fig. 2g). Other hematopoietic growth factors like macrophage colony-stimulating factor (M-CSF; Extended Data Fig. 2e) were detected neither in the supernatant of differentiated bone marrow-derived macrophages (BMDMs) from Plcg2^S707Y/+ mice nor in the plasma of Plcg2^S707Y/+ mice. Multiplex cytokine assessment of skin and lung lysates also revealed increased G-CSF and MIP-1α/MIP-1β in Plcg2^S707Y/+mice, but not in Plcg2^S707Y/+IL-6^−/−, Plcg2^S707Y/+TNF^−/− and Plcg2^S707Y/+caspase-1^−/− mice (Extended Data Fig. 2f,g). In addition, we performed bulk RNA sequencing (RNA-seq) of neutrophils from Plcg2^S707Y/+ mice, which revealed a gene expression signature that closely resembles that of cells treated with G-CSF (Extended Data Fig. 2h)²¹. Taken together, autoinflammation in APLAID was not solely dependent on IL-6, inflammasome or TNF; however, G-CSF production was clearly dysregulated.

**Fig. 2: Minimal rescue of APLAID due to deletion of IL-6, caspase-1 or TNF.**

G-CSF serum levels are high in individuals with APLAID

We examined if the elevated G-CSF levels in Plcg2^S707Y/+ mice are also seen in individuals with APLAID and found increased G-CSF, IL-1β and IL-12p70 in the sera of two treated individuals with APLAID (Fig. 3a)⁵. However, high G-CSF levels were not specific for APLAID and also present in individuals with other monogenic autoinflammatory disease (Extended Data Fig. 3). Participant 1 carries the heterozygous PLCG2 gene mutation p.Ala708Pro (Fig. 3b) and presented with cutaneous lesions at disease onset and mild hypogammaglobulinemia followed by recurrent infections at later stages. Cutaneous flares were well controlled with topical steroids, and mild immunodeficiency was treated with intravenous immunoglobulins. Increased G-CSF levels correlated with increased C-reactive protein (CRP) and disease activity (Fig. 3c). Participant 2 has a compound heterozygous p.Leu845_Leu848 deletion in PLCG2 (Fig. 3b) and clinically suffered from skin rash and recurrent lung bacterial infections leading to bronchiectasis and acute episodes of hemoptysis requiring partial pneumonectomy. Symptoms were inadequately controlled with anakinra, and there was persistent elevation of serum G-CSF (Fig. 3d). B cell immunodeficiency developed in early infancy and did not improve despite the continuous application of intravenous immunoglobulins. These data suggest that although the elevated G-CSF found in participants with APLAID fluctuates over time, it is not simply downstream of IL-1β activation, as observed in other conditions²².

**Fig. 3: High serum levels of G-CSF in participants with APLAID.**

Inhibition of G-CSF reverses established disease in Plcg2 ^S707Y/+ mice

The results from Plcg2^S707Y/+ mice and APLAID participants prompted us to assess the in vivo therapeutic potential of anti-G-CSF treatment. Monoclonal G-CSF antibody (50 µg) was injected intraperitoneally (i.p.) three times per week (Extended Data Fig. 4a) starting 14 d after birth, at which point mice had marked skin inflammation. A matched group of mutant mice received an appropriate IgG1 isotype control i.p. Remarkably, neutralization of G-CSF completely reversed established autoinflammatory disease, reflected by an APLAID skin score of 0 after 4 weeks of treatment (Fig. 4a). Moreover, Plcg2^S707Y/+ mice that received anti-G-CSF injections displayed steady weight gain over time (Fig. 4b) and splenomegaly resolved (Fig. 4c). The lifespan of Plcg2^S707Y/+ mice following G-CSF antagonism was also prolonged (Fig. 4d). ADVIA analysis revealed reduced neutrophil and increased lymphocyte counts in Plcg2^S707Y/+ mice following anti-G-CSF injections (Fig. 4e). Importantly, neutrophil numbers of Plcg2^S707Y/+ mice did not drop below neutrophil counts of Plcg2^+/+ controls. Lymphocyte numbers also recovered in BM, but not in the spleen (Extended Data Fig. 4b), and immunoglobulin subtypes (Ig2a and IgM) decreased in Plcg2^S707Y/+ mice following anti-G-CSF treatment compared to isotype control (Fig. 4f). Thrombopoiesis and erythropoiesis remained largely unaltered in anti-G-CSF-treated Plcg2^S707Y/+ mice (Extended Data Fig. 4c–e) with the exception of the proportion of reticulocytes. FACS analysis of anti-G-CSF-treated Plcg2^S707Y/+ mice showed decreased total leukocytes and myeloid cells, particularly of neutrophils (CD45⁺Ly6G⁺CD11b⁺) across all organs including the spleen, skin and lung (Fig. 4g–i). Histologic examination of isotype control-treated mice revealed persistence of immune cell infiltrates (H&E, MPO⁺ and F4/80⁺) in skin, lung and gut (Fig. 4j) compared to anti-G-CSF-treated mice (Fig. 4k). Thus, anti-G-CSF treatment of Plcg2^S707Y/+ mice rescued the clinical and laboratory features of established APLAID, improving weight gain, spleen size, lifespan and B cell abnormalities.

**Fig. 4: Blocking G-CSF reverses established disease in *Plcg2*^S707Y/+ mice.**

Anti-G-CSF treatment supresses excessive myelopoiesis in Plcg2 ^S707Y/+ mice

Given the high levels of G-CSF together with myeloid cell accumulation in the BM, spleen, skin and lung, we next quantified myeloid progenitors and cell proliferation rates in Plcg2^S707Y/+ mice by a BrdU uptake assay. Plcg2^S707Y/+ mice had a marked increase in the proportion and numbers of neutrophils, monocytes and their progenitors as well as their proliferation rate in the BM (Fig. 5a–c). Notably, the enlarged spleens of Plcg2^S707Y/+ mice also harbored expanded numbers of neutrophils, monocytes and their progenitors and higher frequency of proliferating BrdU⁺ cells (Fig. 5d–g), indicative of enhanced extramedullary myelopoiesis. In contrast, neutrophils, monocytes and macrophages in skin, lung and blood were all BrdU⁻ (Extended Data Fig. 5a–f), consistent with these cells being terminally differentiated in peripheral organs. Representative FACS plots showing the gating strategy were used to identify BrdU⁺ neutrophils and monocytes in BM of a Plcg2^S707Y/+ mouse (Extended Data Fig. 5g).

**Fig. 5: Anti-G-CSF treatment suppresses enhanced myelopoiesis in APLAID mice.**

A colony assay revealed that BM cells and splenocytes from APLAID mutant mice generated increased colonies particularly of the myeloid lineage, when cultured with cytokines such as G-CSF in vitro (Extended Data Fig. 6). In sum, excessive neutrophilia in Plcg2^S707Y/+ mice was driven by both BM and splenic extramedullary myelopoiesis.

To evaluate the importance of G-CSF in driving the enhanced myelopoiesis of Plcg2^S707Y/+ mice, we next investigated cell proliferation rates under anti-G-CSF therapy. Anti-G-CSF treatment reduced myeloid counts and decreased proliferation (BrdU⁺ cells) in the BM (Fig. 5h–k) and the spleen (Fig. 5l–o) relative to isotype control-treated APLAID mutant mice and to comparable levels observed in Plcg2^+/+ mice. These were consistent with the reduced disease activity in anti-G-CSF-treated Plcg2^S707Y/+ mice (Fig. 4). Representative flow cytometry plots showing cell numbers and individual frequencies of BrdU⁺ cells from mice receiving anti-G-CSF antibody treatment or isotype control are shown (Fig. 5p). Taken together, these preclinical data indicate that antagonizing G-CSF prevents autoinflammation in APLAID mutant mice by suppressing medullary and extramedullary myelopoiesis.

Bone marrow transplantation normalizes G-CSF and rescues APLAID phenotype

As indicated by quantitative PCR (qPCR; Extended Data Fig. 6a) and RNA-seq data, elevated G-CSF transcription was observed from both immune (myeloid) and non-immune (fibroblasts) cells from APLAID mice. To further address this, we attempted to rescue the Plcg2^S707Y/+ mediated autoinflammatory disease by transplanting lethally irradiated Plcg2^S707Y/+ mice with PLCG2^+/+ BM (Fig. 6a). Whole BM transplantation of WT cells diminished macroscopic skin inflammation in Plcg2^S707Y/+ mice (Fig. 6b), which was further underscored by a steady weight gain (Fig. 6c), normalized spleen size (Fig. 6d) and prolonged survival rates (Fig. 6e). ADVIA analysis revealed reduced neutrophil/monocyte counts and increased lymphocyte counts in Plcg2^S707Y/+ mice receiving WT BM cells (Fig. 6f). FACS analysis further confirmed decreased myeloid cell counts and no accumulation of neutrophils and monocyte/macrophages in the skin (Fig. 6g–i). BM chimeras exhibited normal amounts of G-CSF in plasma (Fig. 6j). Tissue pathology in Plcg2^S707Y/+ mice was prevented in BM chimeras with only a few immune cell infiltrates in skin and lung (Fig. 6k).

**Fig. 6: Bone marrow transplantation normalizes G-CSF levels and rescues phenotype in *Plcg2*^S707Y/+ mice.**

Importantly, irradiation alone did not improve signs of skin autoinflammation in Plcg2^S707Y/+ mice receiving BM from Plcg2^S707Y/+ mice (Extended Data Fig. 7b,c). Irradiated Plcg2^S707Y/+ mice still displayed weight loss (Extended Data Fig. 7d), splenomegaly (Extended Data Fig. 7e), reduced survival (Extended Data Fig. 7f) and neutrophilia (Extended Data Fig. 7g), while thrombopoiesis and erythropoiesis remained unaltered as determined by ADVIA analysis (Extended Data Fig. 7h–j). Donor WT CD45.1⁺ cells fully reconstituted immune cells in the blood and spleen at 7 weeks after transplantation (Extended Data Fig. 7k) in Plcg2^S707Y/+ mice. FACS analysis of monocytes/macrophages confirmed a high degree of reconstitution with healthy WT donor cells in the lung, blood, spleen and BM, while there were less than 50% of skin macrophages (that is, Langerhans cells) in donor-derived WT CD45.1⁺ cells (Extended Data Fig. 7k). This is consistent with the embryonic origin of Langerhans cells and their high resistance to irradiation²³.

These BM chimera experiments conclude that the host can still remain healthy with a PLCG2 mutation in all non-hematopoietic cells. However, even though the PLCG2 mutation is required in a hematopoietic cell, this does not establish if G-CSF is intrinsically coming from the same cell type.

Delayed onset of APLAID phenotype in G-CSF-deficient mice

Next, we transferred Plcg2^S707Y/+ BM cells into either irradiated WT mice or G-CSF^−/− mice (Fig. 7a) to assess if G-CSF originates from hematopoietic or non-hematopoietic cells in our APLAID disease model. As expected, transfer of BM cells from Plcg2^S707Y/+ mice resulted in inflammation as indicated by pronounced weight loss (Fig. 7b,c), reduced survival (Fig. 7d) and elevated G-CSF (Fig. 7e) levels. In contrast, G-CSF knockouts showed a delayed onset of the APLAID phenotype (Fig. 7b–d). In accordance with the onset of clinical signs and body weight loss at day 70 following irradiation and reconstitution, increasing levels of G-CSF were observed in G-CSF^−/− mice receiving Plcg2^S707Y/+ BM (Fig. 7e). Also, elevated IL-12 (p40) and eotaxin levels were observed while animals are still asymptomatic (Extended Data Fig. 8a). Of note, the BM chimera with the highest G-CSF level was the one that had the most severe weight loss.

ADVIA analysis at day 18 showed APLAID-specific blood count abnormalities (increased neutrophils/monocytes and decreased numbers of lymphocytes; Fig. 7f) in irradiated and reconstituted WT mice, while thrombopoiesis and erythropoiesis (Fig. 7g,h) remained unaltered except for increased reticulocyte counts (Fig. 7i). Taken together, these findings suggest that G-CSF is predominantly derived from radio-resistant cells in our model, but it can eventually be made by hematopoietic cells too. Importantly, our data argue in favor of allogeneic BM transplantation or G-CSF neutralization as viable treatment options for individuals with APLAID.

Discussion

Using genetic and pharmacological approaches, we identify G-CSF as the key driver of autoinflammation in APLAID caused by missense mutations in PLCG2. Allogeneic BM transplantation may be beneficial in individuals with APLAID as indicated by our preclinical dataset. Normalized G-CSF levels and reversal of tissue pathology in BM chimeras further underscore the pivotal role of this cytokine in perpetuating autoinflammation in APLAID. Moreover, the observation that Plcg2^S707Y/+ mice with established disease can be completely rescued by anti-G-CSF treatment, without inducing neutropenia, may offer a more appealing therapeutic strategy for individuals with APLAID in the future.

In the beginning, we undertook a genetic approach to define the molecular mechanisms and signaling pathways in APLAID. Although IL-6, inflammasome and TNF signaling have been implicated by previous in vitro studies¹⁸, our work demonstrates that disease in Plcg2^S707Y/+ mice is largely independent of these factors. In contrast, we identify a specific role for enhanced G-CSF signaling in causing autoinflammatory disease in Plcg2^S707Y/+ mice. Although G-CSF levels are profoundly elevated in APLAID mice, this increase was not specific to APLAID participants compared to other monogenic autoinflammatory diseases. However, G-CSF levels in individuals with APLAID remained elevated despite the administration of biologics targeting IL-1, for example, arguing that G-CSF is not a nonspecific feature of inflammation in APLAID. Furthermore, our results explain the inadequate treatment responses to IL-1 or TNF blockade in individuals with APLAID.

Increased G-CSF levels across various autoinflammatory diseases levels may be the result of a shared organ pathology characterized by predominantly neutrophilic infiltrates. However, it is unknown if anti-G-CSF therapy might also be beneficial in hyper-IgD with periodic fever syndrome, deficiency of adenosine deaminase-2 or other autoinflammatory conditions. Although neutralization of G-CSF or its receptor has shown to be a promising therapeutic target in inflammatory joint diseases such as rheumatoid arthritis^24,25 and other autoimmune disorders^26,27, we are unaware of preclinical studies or patient reports in autoinflammatory diseases. Our preclinical observations alongside G-CSF elevation in other autoinflammatory diseases implicate that targeting G-CSF can be considered in a variety of neutrophil-associated, monogenetically inherited human inflammatory diseases and could determine whether G-CSF is the primary driver in these autoinflammatory conditions or merely a disease modifier in a complex inflammatory cascade.

Importantly, defects in lymphopoiesis were restored in Plcg2^S707Y/+ mice receiving anti-G-CSF treatment. A possible explanation is that elevated G-CSF can suppress lymphopoiesis by targeting stromal cells that contribute to lymphoid niches in the BM²⁸. This is a relevant consideration in determining the benefit of anti-G-CSF treatment for individuals with APLAID, where immunosuppression of neutrophil function is a potential concern. However, given that targeting G-CSF did not cause neutropenia and may actually resolve some underlying immunodeficiency, this is mitigated to some extent.

G-CSF has various hematological effects including the activation of endothelial cells²⁹, modulation of the leukocyte adhesion molecule expression^30,31, angiogenesis²⁰ and chemokine production³⁰. Additionally, G-CSF also enhances neutrophil and macrophage phagocytosis^32,33 and prolongs neutrophil survival³⁴. In agreement with this, we found enhanced myelopoiesis in the BM and also in extramedullary organs like the spleen. Therefore, in APLAID, overproduction of G-CSF might act both locally by mediating trafficking of myeloid cells through the endothelium, as well as promoting local cellular survival within inflamed sites and systemically via increased proliferation of hematopoietic progenitor cells in the BM and spleen and the release of mature neutrophils into the blood. Interestingly, while G-CSF levels in tissue lysates were undetectable, they remained elevated in plasma samples of Plcg2^S707Y/+IL-6^−/−, Plcg2^S707Y/+caspase-1^−/− and Plcg2^S707Y/+TNF^−/− mice, highlighting the superior role of systemic effects over its contribution to the local milieu in our APLAID model. Furthermore, these data point toward a close interplay between G-CSF and each of these proinflammatory cytokine pathways and its multistep regulation of inflammasome and NF-kB activity, which is likely to be complex and multifactorial.

Nonetheless, the results from chimeras, where G-CSF was deleted from the non-hematopoietic compartment clearly demonstrate that this is where that cytokine is produced, in response to PLCG2 mutant BM. Therefore, there should be a secreted factor, metabolite, or other cellular interaction, whereby the mutant PLCG2 hematopoietic cell signals to the G-CSF-producing non-hematopoietic cell (Extended Data Fig. 8b).

PLCγ2 has an established role downstream of the B cell receptor^11,35 and other cell surface receptors^36,37,38, driving the conversion of PIP₂ to DAG, which can activate PKC and subsequent transcriptional responses^39,40. However, the shared biological pathway between PLCγ2 and G-CSF remains unknown. Further light has been shed on this issue by in vivo studies demonstrating that PKC activation can trigger a mouse model of inflammatory skin disease due to G-CSF⁴¹. Furthermore, PLCγ2 expression in neutrophils was found to be required for the development of arthritis in the K/BxN mouse model⁴², which is also known to be G-CSF dependent⁴³. Interestingly, the cell-permeable DAG analog 1-oleoyl-2-acetyl-sn-glycerol has been observed to drive G-CSF transcription in human monocytes in vitro⁴⁴. A soluble factor such as this could be a good candidate to signal from mutant PLCG2 BM cells to trigger G-CSF transcription from a non-hematopoietic cell, as required in this model of APLAID (Extended Data Fig. 8b); however, this is highly speculative. Although further confirmation is required, we suggest this mechanism could account for increased G-CSF production due to the gain-of-function PLCG2 mutations causing APLAID.

In summary, by using a Plcg2^S707Y/+ mouse model, we identified a G-CSF-driven monogenic autoinflammatory disease. Treatment blocking G-CSF can reverse established autoinflammatory disease and restore B cells. Based on this preclinical model of APLAID, patients may benefit from allogeneic BM transplantation or neutralization of G-CSF.

Methods

Generation of the Plcg2 ^S707Y/+ mouse model

The point mutation p.Ser707Tyr in PLCG2 causes APLAID¹. The presence of a Neomycin cassette prevents the expression of the knock-in point mutation. Only Flp-mediated excision enables the deletion of FRT-flanked Neomycin cassette, thereby generating the inflammatory phenotype. This deletion was performed in vivo, by breeding heterozygous PLCγ2 offspring harboring the recombined allele with C57BL/6 Flp deleter mice expressing ubiquitous Flp-recombinase⁴⁵. The Plcg2^S707Y/+ mouse model was established by GSK.

Mice

The following mouse strains were used: Plcg2^S707Y/+IL-6^−/− (ref. ⁴⁶), Plcg2^S707Y/+caspase-1^−/− (ref. ⁴⁷), Plcg2^S707Y/+TNF^−/− (ref. ⁴⁸), G-CSF^−/− (ref. ⁴⁹) and Ly5.1/J (ref. ⁵⁰). All genetically modified mouse strains in this study were on the C57BL/6J background, bred and maintained at the Walter and Eliza Hall Institute of Medical Research. Randomization was applied to mouse sex. Both male and female Plcg2^S707Y/+mice at 26 weeks of age and their littermate controls were used for all experiments. Cohort sizes are described in each figure legend. All animal studies were ethically reviewed by and carried out in accordance with approval from the Walter and Eliza Hall Animal Committee (2017.029 and 2020.017).

Genotyping

All genotyping of mouse strains was PCR based. Genomic tail DNA was obtained using DirectPCR Lysis Reagent (Viagen Biotech) per the manufacturer’s instructions. PCR products were run on 1.5–2% agarose gels and visualized using ethidium bromide staining. Primers were purchased from Integrated DNA Technologies, and oligonucleotide details are provided in Supplementary Table 1.

Clinical scoring

Clinical assessment for skin inflammation used the following scoring system: 0, no clinical signs of skin inflammation; 1, early signs of inflammation including flaky skin and scruffy fur; 2, skin lesions occur in one body region including paws or ears or tail and appear dry/not open; 3, skin lesions appear in two body regions (including paws and/or tail and/or ear) and present dry/not open; 4, skin lesions appear in three regions (including paws and/or tail and/or ear) and present dry/not open; 5, severe, open lesions are present in either of one above-mentioned body regions (paws, ear, tail), >20% loss of weight; scoring was performed in a blinded manner.

Human samples

The study included sera of two affected APLAID individuals from two distinct families. Participant 1 has a heterozygous p.Ala708Pro mutation in PLGC2 gene. In participant 2, the heterozygous mutation p.Leu845_Leu848del in PLGC2 gene was detected. Written informed consent was obtained from all participants and healthy control volunteers. The Ethical Review Board of Hospital Clinic, Barcelona, Spain, approved the study (HCB/2019/0631). All investigations were performed in accordance with the ethical standards of the 1964 Declaration of Helsinki and its later amendments.

Hematology

Automated cell counts were performed on mouse blood collected from the submandibular vein into Microtainer tubes containing EDTA (Sarstedt), using an ADVIA2120i hematological analyzer (Siemens).

Flow cytometry

Cells were isolated after perfusion through the left cardiac ventricle with PBS. Skin tissue including paws, tail, ears and lungs were then incubated in digestion buffer (2 mg ml⁻¹ Collagenase IV (Worthington Biochemical), 1 mg ml⁻¹ Dispase (Worthington Biochemical) and 0.5 mg ml⁻¹ DNase I (Sigma) in PBS)) for 45 min at 37 °C, with agitation. Cells released during digestion were filtered through a 100-μm nylon mesh. Erythrocytes were lysed with red cell lysis buffer (156 mM NH₄Cl, 11.9 mM NaHCO₃, 0.097 mM EDTA). Cells were stained with viability dyes, incubated with FcR block, and stained with fluorochrome-conjugated antibodies (see Supplementary Table 1 for antibodies, clones, fluorochromes and manufacturers). Cells were washed and fixed with Cytofix/Cytoperm buffer (BD Biosciences) for 20 min on ice in the dark, then stored at 4 °C. Quantification of total cell numbers by flow cytometry was done using fluorescent beads (Beckman Coulter). Data were acquired on a BD Symphony A5 flow cytometer. Compensation and analysis of the flow cytometry data were performed using FlowJo software (Tree Star).

BrdU assay

To assess BrdU incorporation into hematopoietic stem and progenitor cells, mice were i.p. injected with a single dose of 1 mg BrdU 2 h before organ collection. Cells were labeled with the indicated FACS antibodies, then fixed and stained with a BrdU antibody using the FITC-BrdU Flow kit (BD Biosciences) according to the manufacturer’s protocol.

Histology

Organs were collected in 10% neutral buffered formalin. Organ sections were prepared from paraffin blocks and stained with H&E. IHC was performed using the antibodies summarized in Supplementary Table 1.

Computed tomography scan

For Micro-CT scans, mice were euthanized and scanned on a Bruker Skyskan 1276 Micro-CT. Images were analyzed using Fiji/BoneJ software.

Multiplex magnetic bead-based immunoassays

Mouse plasma samples were collected and immediately stored at −20 °C. A total of 23 analytes were measured using the Bio-Plex Pro Mouse Cytokine 23-plex Assay. These include IL-1α, IL-1β, IL-2, IL-3, IL-4, IL-5, IL-6, IL-9, IL-10, IL-12 (p40), IL-12 (p70), IL-13, IL-17A, eotaxin, G-CSF, GM-CSF, IFN-γ, KC, MCP-1 (MCAF), MIP-1α/β, RANTES and TNF. The assay plate was pre-coated with beads labeled with fluorescent dye and was washed two times with Bio-Plex Wash Buffer on a Bio-Plex handheld magnetic washer. The plate was sealed with a sealing tape after samples and standards were added and incubated on a microtiter plate shaker for 30 min. After incubation, the plate was washed three times with wash buffer on a handheld magnetic washer. Biotinylated detection antibodies were incubated with the bead–antibody bound samples for 30 min and washed three times with wash buffer. Next, streptavidin conjugated to phycoerythrin was added and incubated for 10 min to allow formation of the final detection complex. Finally, the plate was washed three times, before the complexes were resuspended in assay buffer for acquisition on the Bio-Plex 200 array reader using the Bio-Plex Manager software. Five-parameter logistic function was used to generate the standard curve, in which a regression analysis was performed to report the fluorescence intensity and observed concentration of each analyte.

Human plasma samples were measured using the Milliplex Cytokine/Chemokine/Growth Factor Panel A-Immunology Multiplex Assay, according to manufacturer’s protocol. Catalog numbers and manufacturers’ details of both commercial kits are provided in Supplementary Table 1.

ELISA

All ELISA experiments (G-CSF, M-CSF from R&D and immunoglobulins subtypes from Thermo Fisher Scientific) were carried out according to the manufacturer’s instructions. Catalog numbers and manufacturers’ details are given in Supplementary Table 1.

Colony assay

Single-cell suspensions from BM and spleen were prepared in balanced salt solution (0.15 M NaCl, 4 mM KCl, 2 mM CaCl₂, 1 mM MgSO₄, 1 mM KH₂PO₄, 0.8 mM K₂HPO₄ and 15 mM N-2-hydroxyethylpiperazine-N′-2-ethanesulfonic acid supplemented with 2% (vol/vol) bovine calf serum). Clonal analysis of BM cells (2.5 × 10⁴) and splenocytes (5 × 10⁴) was performed in 1 ml semisolid agar cultures of 0.3% agar in DMEM containing 20% newborn calf serum, stem cell factor (100 ng ml⁻¹; in-house), erythropoietin (2 U ml⁻¹; Janssen), IL-3 (10 ng ml⁻¹; in-house), G-CSF (103 U ml⁻¹; PeproTech), granulocyte–macrophage colony-stimulating factor (103 U ml⁻¹; in-house) and/or M-CSF (103 U ml⁻¹; in-house). Cultures were incubated at 37 °C for 7 d in a fully humidified atmosphere of 10% CO₂ in air, then fixed, dried onto glass slides and stained for acetylcholinesterase, Luxol fast blue, hematoxylin and the number and type of colonies were determined.

G-CSF antibody treatment

Mice were injected i.p. with an anti-G-CSF rat IgG1 monoclonal antibody (clone MAB414, 50 μg per injection, R&D) to neutralize G-CSF or an IgG1 isotype control (clone MAB005, 50 μg per injection, R&D). Antibody injections were given three times a week starting on day 14 after birth for a total of four consecutive weeks.

Bone marrow chimeras

BM was flushed from the femurs and filtered through a 100-µm filter. A total of 1 × 10⁶ cells in 200 µl PBS were intravenously injected into lethally irradiated (2 × 550 rads) mice.

Cell sorting

FACS sorting of primary mouse cells was performed by staff of the Walter and Eliza Hall Institute of Medical Research flow cytometry core facility (Melbourne, Australia). Depending on the experimental requirements and instrument availability, the BD FACSAria W (lasers: 375 nm, 405 nm, 488 nm, 561 nm and 640 nm) and/or BD FACSAria Fusion (lasers: 405 nm, 488 nm, 561 nm and 640 nm; BD Biosciences) were used. Before sorting, cells were resuspended in 1× PBS/2% FCS and kept on ice. Neutrophils (CD45⁺, CD11b⁺, Ly6G⁺, F4/80⁻, CD115⁻, CD64⁻), macrophages (CD45⁺, CD11b⁺, F4/80⁺, CD115⁺, CD64⁺, Ly6G⁻) and stromal cells (endothelial cells: CD45⁻, CD31⁺; fibroblasts: CD45⁻, CD31⁻, gp38⁺ PDGFRa⁺; keratinocytes: CD45⁻, CD31⁻, PDGFRa⁻, CD49f⁺; and epithelial cells: CD45⁻ CD31⁻, PDGFRa⁻, EPCAM⁺) were directly sorted into 1.5-ml microcentrifuge tubes containing RLT lysis buffer (RNeasy Plus Micro Kit; QIAGEN) using a 70-μm or 100-μm nozzle as required.

RNA isolation and quantitative real-time PCR

Total RNA was extracted from sorted cells using an RNeasy Plus Micro Kit (QIAGEN) and reverse-transcribed into complementary DNA using SuperScript III reverse transcription and oligo(dT) nucleotides (Thermo Fisher Scientific). Real-time qPCR was performed using SYBR Green/ROX qPCR Master Mix (Thermo Fisher Scientific) on a ViiA 7 PCR System (Thermo Fisher Scientific). Cytokine expression from qPCR was calculated as gene expression relative to gapdh as ΔCT. Primers for the genes assessed are shown in Supplementary Table 1.

RNA-seq

The sequencing workflow was performed as previously described⁵¹. Briefly, RNA from sorted neutrophils and macrophages (purity 99–100%) was extracted with an RNeasy Plus Micro Kit (QIAGEN) according to the manufacturer’s protocol. Extracted RNA quality and quantity were assessed using an Agilent 2200 TapeStation System (Agilent) with RNA ScreenTapes (Agilent). Next-generation sequencing libraries were created with 10 ng of RNA from samples with distinct 18S and 28S peaks and RNA integrity number values greater than 7, using an NEBNext Ultra II Directional RNA Library Prep Kit from Illumina (New England Biolabs) according to the manufacturer’s protocol. RNA libraries were pooled and sent for single-end 75-bp sequencing at the Genomics Laboratory of The Walter and Eliza Hall Institute of Medical Research on a NextSeq 500 next-generation sequencer (Illumina) to obtain ∼20 million reads per sample.

Bioinformatics

Paired-end 75-bp RNA-seq short reads were generated using a NextSeq 500 (Illumina). Between 16 and 64 million read pairs were generated for each sample, and reads were aligned to the Mus musculus genome (GRCm39/mm39) using Rsubread⁵². The number of read pairs overlapping mouse genes was summarized using featureCounts and Gencode (vM27) annotation. Lowly expressed genes were filtered out using edgeR’s filterByExpr function⁵³. Sex- linked genes and genes without current annotation were also filtered. Differential expression analysis was undertaken using the edgeR and limma⁵⁴ software packages. Library sizes were normalized using the quantile normalization method⁵⁵ using limma’s normalizeBetweenArrays function. Sample relative quality weights were estimated using arrayWeights⁵⁶, and differential expression was evaluated using limma trend⁵⁷ with robust empirical Bayes estimation of the variances⁵⁸. The effect caused by age differences of the mice was adjusted and correlations between repeated measurements from the same mouse were estimated using the duplicateCorrelation method⁵⁹. The false discovery rate was controlled below 0.1 using the method of Benjamini and Hochberg. Overrepresentation of Gene Ontology terms and KEGG pathways for the differentially expressed genes were identified using limma’s goana and kegga functions. Heat maps were drawn using the pheatmap function. Macrophage samples and neutrophil samples were analyzed separately. Enrichment of gene sets²¹ was tested using the roast method⁶⁰ and illustrated using barcode plots drawn by limma’s barcodeplot function.

Statistics and reproducibility

No statistical methods were used to predetermine sample sizes, but our sample sizes are similar to those reported in previous publications⁶¹. Data distribution was assumed to be normal, but this was not formally tested. Differences in clinical score between the groups were assessed using two-way ANOVA with Bonferroni post-test correction for repeated measurements. For other experiments, differences between groups were determined using unpaired or paired two-sided Student’s t-tests as indicated. The statistical test for the survival curve was determined by a Mantel–Cox test. No data points were excluded from the analyses.

All analyses were performed using GraphPad Prism v9.4.1. Each dot represents one mouse or one individual. Data are shown as means ± s.e.m. unless stated otherwise. Confidence intervals were set to 95%. Differences were statistically significant when P values were less than 0.05.

Reporting summary

Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.

Data availability

RNA-seq data have been deposited in the Gene Expression Omnibus under accession https://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc=GSE211109. Biological materials that are not commercially available can be requested from the authors under a material transfer agreement. Source data are provided with this paper.

References

Zhou, Q. et al. A hypermorphic missense mutation in PLCG2, encoding phospholipase Cγ2, causes a dominantly inherited autoinflammatory disease with immunodeficiency. Am. J. Hum. Genet. 91, 713–720 (2012).
Article CAS PubMed PubMed Central Google Scholar
Neves, J. F. et al. Novel PLCG2 mutation in a patient with APLAID and cutis laxa. Front. Immunol. 9, 2863 (2018).
Article CAS PubMed PubMed Central Google Scholar
Novice, T. et al. A germline mutation in the C2 domain of PLCγ2-associated with gain-of-function expands the phenotype for PLCG2-related diseases. J. Clin. Immunol. 40, 267–276 (2020).
Article CAS PubMed Google Scholar
Moran-Villasenor, E. et al. Expanding the clinical features of autoinflammation and phospholipase Cγ2-associated antibody deficiency and immune dysregulation by description of a novel patient. J. Eur. Acad. Dermatol. Venereol. 33, 2334–2339 (2019).
Article CAS PubMed Google Scholar
Martin-Nalda, A. et al. Severe autoinflammatory manifestations and antibody deficiency due to novel hypermorphic PLCG2 mutations. J. Clin. Immunol. 40, 987–1000 (2020).
Article CAS PubMed PubMed Central Google Scholar
Park, H. S. et al. A novel likely pathogenic PLCG2 variant in a patient with a recurrent skin blistering disease and B cell lymphopenia. Eur. J. Med. Genet. 65, 104387 (2022).
Article CAS PubMed Google Scholar
Ombrello, M. J. et al. Cold urticaria, immunodeficiency, and autoimmunity related to PLCG2 deletions. N. Engl. J. Med. 366, 330–338 (2012).
Article CAS PubMed PubMed Central Google Scholar
Burger, J. A. et al. Clonal evolution in patients with chronic lymphocytic leukaemia developing resistance to BTK inhibition. Nat. Commun. 7, 11589 (2016).
Article CAS PubMed PubMed Central Google Scholar
Sims, R. et al. Rare coding variants in PLCG2, ABI3, and TREM2 implicate microglial-mediated innate immunity in Alzheimer’s disease. Nat. Genet. 49, 1373–1384 (2017).
Article CAS PubMed PubMed Central Google Scholar
Liu, Y. et al. Structural insights and activating mutations in diverse pathologies define mechanisms of deregulation for phospholipase C gamma enzymes. EBioMedicine 51, 102607 (2020).
Article PubMed PubMed Central Google Scholar
Kim, Y. J., Sekiya, F., Poulin, B., Bae, Y. S. & Rhee, S. G. Mechanism of B cell receptor-induced phosphorylation and activation of phospholipase C-γ2. Mol. Cell. Biol. 24, 9986–9999 (2004).
Article CAS PubMed PubMed Central Google Scholar
Coggeshall, K. M., McHugh, J. C. & Altman, A. Predominant expression and activation-induced tyrosine phosphorylation of phospholipase C-γ2 in B lymphocytes. Proc. Natl Acad. Sci. USA 89, 5660–5664 (1992).
Article CAS PubMed PubMed Central Google Scholar
Jackson, J. T., Mulazzani, E., Nutt, S. L. & Masters, S. L. The role of PLCγ2 in immunological disorders, cancer, and neurodegeneration. J. Biol. Chem. 297, 100905 (2021).
Article CAS PubMed PubMed Central Google Scholar
Everett, K. L. et al. Characterization of phospholipase C gamma enzymes with gain-of-function mutations. J. Biol. Chem. 284, 23083–23093 (2009).
Article CAS PubMed PubMed Central Google Scholar
Engelke, M. et al. Cutting edge: feed-forward activation of phospholipase Cγ2 via C2 domain-mediated binding to SLP65. J. Immunol. 191, 5354–5358 (2013).
Article CAS PubMed Google Scholar
Obba, S. et al. The PRKAA1/AMPKα1 pathway triggers autophagy during CSF1-induced human monocyte differentiation and is a potential target in CMML. Autophagy 11, 1114–1129 (2015).
Article CAS PubMed PubMed Central Google Scholar
Leon, C. M. et al. Requirement for PLCγ2 in IL-3 and GM-CSF-stimulated MEK/ERK phosphorylation in murine and human hematopoietic stem/progenitor cells. J. Cell. Physiol. 226, 1780–1792 (2011).
Article CAS PubMed Google Scholar
Chae, J. J. et al. Connecting two pathways through Ca²⁺ signaling: NLRP3 inflammasome activation induced by a hypermorphic PLCG2 mutation. Arthritis Rheumatol. 67, 563–567 (2015).
Article CAS PubMed PubMed Central Google Scholar
Yu, P. et al. Autoimmunity and inflammation due to a gain-of-function mutation in phospholipase Cγ2 that specifically increases external Ca²⁺ entry. Immunity 22, 451–465 (2005).
Article PubMed Google Scholar
Abe, K. et al. A novel N-ethyl-N-nitrosourea-induced mutation in phospholipase Cγ2 causes inflammatory arthritis, metabolic defects, and male infertility in vitro in a murine model. Arthritis Rheum. 63, 1301–1311 (2011).
Article CAS PubMed Google Scholar
Avila-Portillo, L. M. et al. Comparative analysis of the biosimilar and innovative G-CSF modulated pathways on umbilical cord blood-derived mononuclear cells. Bioinform. Biol. Insights 14, 1177932220913307 (2020).
Article CAS PubMed PubMed Central Google Scholar
Zong, M. et al. Anakinra treatment in patients with refractory inflammatory myopathies and possible predictive response biomarkers: a mechanistic study with 12 months follow-up. Ann. Rheum. Dis. 73, 913–920 (2014).
Article CAS PubMed Google Scholar
Merad, M. et al. Langerhans cells renew in the skin throughout life under steady-state conditions. Nat. Immunol. 3, 1135–1141 (2002).
Article CAS PubMed PubMed Central Google Scholar
Lawlor, K. E. et al. Critical role for granulocyte colony-stimulating factor in inflammatory arthritis. Proc. Natl Acad. Sci. USA 101, 11398–11403 (2004).
Article CAS PubMed PubMed Central Google Scholar
Campbell, I. K. et al. Therapeutic targeting of the G-CSF receptor reduces neutrophil trafficking and joint inflammation in antibody-mediated inflammatory arthritis. J. Immunol. 197, 4392–4402 (2016).
Article CAS PubMed Google Scholar
Goldberg, G. L. et al. G-CSF and neutrophils are nonredundant mediators of murine experimental autoimmune uveoretinitis. Am. J. Pathol. 186, 172–184 (2016).
Article CAS PubMed Google Scholar
Rumble, J. M. et al. Neutrophil-related factors as biomarkers in EAE and MS. J. Exp. Med. 212, 23–35 (2015).
Article CAS PubMed PubMed Central Google Scholar
Day, R. B., Bhattacharya, D., Nagasawa, T. & Link, D. C. Granulocyte colony-stimulating factor reprograms bone marrow stromal cells to actively suppress B lymphopoiesis in mice. Blood 125, 3114–3117 (2015).
Article CAS PubMed PubMed Central Google Scholar
Bussolino, F. et al. Granulocyte- and granulocyte-macrophage-colony stimulating factors induce human endothelial cells to migrate and proliferate. Nature 337, 471–473 (1989).
Article CAS PubMed Google Scholar
Suzuki, S. et al. Autocrine production of epithelial cell-derived neutrophil attractant-78 induced by granulocyte colony-stimulating factor in neutrophils. Blood 99, 1863–1865 (2002).
Article PubMed Google Scholar
Sugimori, N. et al. Administration of G−CSF to normal individuals diminishes L-selectin⁺ T cells in the peripheral blood that respond better to alloantigen stimulation than L-selectin- T cells. Bone Marrow Transplant. 23, 119–124 (1999).
Article CAS PubMed Google Scholar
Fattorossi, A. et al. Effects of granulocyte colony-stimulating factor and granulocyte/macrophage colony-stimulating factor administration on T cell proliferation and phagocyte cell-surface molecules during hematopoietic reconstitution after autologous peripheral blood progenitor cell transplantation. Cancer Immunol. Immunother. 49, 641–648 (2001).
Article CAS PubMed Google Scholar
Bermudez, L. E., Petrofsky, M. & Stevens, P. Treatment with recombinant granulocyte colony-stimulating factor (Filgrastin) stimulates neutrophils and tissue macrophages and induces an effective non-specific response against Mycobacterium avium in mice. Immunology 94, 297–303 (1998).
Article CAS PubMed PubMed Central Google Scholar
Villunger, A., Scott, C., Bouillet, P. & Strasser, A. Essential role for the BH3-only protein Bim but redundant roles for Bax, Bcl-2 and Bcl-w in the control of granulocyte survival. Blood 101, 2393–2400 (2003).
Article CAS PubMed Google Scholar
Benschop, R. J., Brandl, E., Chan, A. C. & Cambier, J. C. Unique signaling properties of B cell antigen receptor in mature and immature B cells: implications for tolerance and activation. J. Immunol. 167, 4172–4179 (2001).
Article CAS PubMed Google Scholar
Moreno-Garcia, M. E. et al. CD38 signaling regulates B lymphocyte activation via a phospholipase C (PLC)-γ2-independent, protein kinase C, phosphatidylcholine-PLC, and phospholipase D-dependent signaling cascade. J. Immunol. 174, 2687–2695 (2005).
Article CAS PubMed Google Scholar
Ying, H., Li, Z., Yang, L. & Zhang, J. Syk mediates BCR- and CD40-signaling integration during B cell activation. Immunobiology 216, 566–570 (2011).
Article CAS PubMed Google Scholar
Bai, L. et al. Phospholipase Cγ2 contributes to light-chain gene activation and receptor editing. Mol. Cell. Biol. 27, 5957–5967 (2007).
Article CAS PubMed PubMed Central Google Scholar
de Gorter, D. J., Vos, J. C., Pals, S. T. & Spaargaren, M. The B cell antigen receptor controls AP-1 and NFAT activity through Ras-mediated activation of Ral. J. Immunol. 178, 1405–1414 (2007).
Article PubMed Google Scholar
Wang, J. H. et al. Selective defects in the development of the fetal and adult lymphoid system in mice with an Ikaros null mutation. Immunity 5, 537–549 (1996).
Article CAS PubMed Google Scholar
Cataisson, C. et al. CXCR2 ligands and G-CSF mediate PKCα-induced intraepidermal inflammation. J. Clin. Invest. 116, 2757–2766 (2006).
Article CAS PubMed PubMed Central Google Scholar
Futosi, K., Kasa, O., Szilveszter, K. P. & Mocsai, A. Neutrophil phospholipase Cγ2 drives autoantibody-induced arthritis through the generation of the inflammatory microenvironment. Arthritis Rheumatol. 73, 1614–1625 (2021).
Article CAS PubMed Google Scholar
Christensen, A. D., Haase, C., Cook, A. D. & Hamilton, J. A. Granulocyte colony-stimulating factor (G-CSF) plays an important role in immune complex-mediated arthritis. Eur. J. Immunol. 46, 1235–1245 (2016).
Article CAS PubMed Google Scholar
Vellenga, E., van der Vinne, B., De Wolf, J. T. & Halie, M. R. Simultaneous expression and regulation of G-CSF and IL-6 mRNA in adherent human monocytes and fibroblasts. Br. J. Haematol. 78, 14–18 (1991).
Article CAS PubMed Google Scholar
Rodriguez, C. I. et al. High-efficiency deleter mice show that FLPe is an alternative to Cre-loxP. Nat. Genet. 25, 139–140 (2000).
Article CAS PubMed Google Scholar
Kopf, M. et al. Impaired immune and acute-phase responses in interleukin-6-deficient mice. Nature 368, 339–342 (1994).
Article CAS PubMed Google Scholar
Schott, W. H. et al. Caspase-1 is not required for type 1 diabetes in the NOD mouse. Diabetes 53, 99–104 (2004).
Article CAS PubMed Google Scholar
Korner, H., Lemckert, F. A., Chaudhri, G., Etteldorf, S. & Sedgwick, J. D. Tumor necrosis factor blockade in actively induced experimental autoimmune encephalomyelitis prevents clinical disease despite activated T cell infiltration to the central nervous system. Eur. J. Immunol. 27, 1973–1981 (1997).
Article CAS PubMed Google Scholar
Lieschke, G. J. et al. Mice lacking granulocyte colony-stimulating factor have chronic neutropenia, granulocyte and macrophage progenitor cell deficiency, and impaired neutrophil mobilization. Blood 84, 1737–1746 (1994).
Article CAS PubMed Google Scholar
Schluns, K. S., Williams, K., Ma, A., Zheng, X. X. & Lefrancois, L. Cutting edge: requirement for IL-15 in the generation of primary and memory antigen-specific CD8 T cells. J. Immunol. 168, 4827–4831 (2002).
Article CAS PubMed Google Scholar
Louis, C. et al. NK cell-derived GM-CSF potentiates inflammatory arthritis and is negatively regulated by CIS. J. Exp. Med. 217, e20191421 (2020).
Article CAS PubMed PubMed Central Google Scholar
Liao, Y., Smyth, G. K. & Shi, W. The R package Rsubread is easier, faster, cheaper and better for alignment and quantification of RNA sequencing reads. Nucleic Acids Res. 47, e47 (2019).
Article CAS PubMed PubMed Central Google Scholar
Chen, Y., Lun, A. T. & Smyth, G. K. From reads to genes to pathways: differential expression analysis of RNA-seq experiments using Rsubread and the edgeR quasi-likelihood pipeline. F1000Res 5, 1438 (2016).
PubMed PubMed Central Google Scholar
Ritchie, M. E. et al. limma powers differential expression analyses for RNA-sequencing and microarray studies. Nucleic Acids Res. 43, e47 (2015).
Article PubMed PubMed Central Google Scholar
Smyth, G. K. & Speed, T. Normalization of cDNA microarray data. Methods 31, 265–273 (2003).
Article CAS PubMed Google Scholar
Ritchie, M. E. et al. Empirical array quality weights in the analysis of microarray data. BMC Bioinformatics 7, 261 (2006).
Article PubMed PubMed Central Google Scholar
Law, C. W., Chen, Y., Shi, W. & Smyth, G. K. voom: Precision weights unlock linear model analysis tools for RNA-seq read counts. Genome Biol. 15, R29 (2014).
Article PubMed PubMed Central Google Scholar
Phipson, B., Lee, S., Majewski, I. J., Alexander, W. S. & Smyth, G. K. Robust hyperparameter estimation protects against hypervariable genes and improves power to detect differential expression. Ann. Appl. Stat. 10, 946–963 (2016).
Article PubMed PubMed Central Google Scholar
Smyth, G. K., Michaud, J. & Scott, H. S. Use of within-array replicate spots for assessing differential expression in microarray experiments. Bioinformatics 21, 2067–2075 (2005).
Article CAS PubMed Google Scholar
Wu, D. et al. ROAST: rotation gene-set tests for complex microarray experiments. Bioinformatics 26, 2176–2182 (2010).
Article CAS PubMed PubMed Central Google Scholar
Brydges, S. D. et al. Inflammasome-mediated disease animal models reveal roles for innate but not adaptive immunity. Immunity 30, 875–887 (2009).
Article CAS PubMed PubMed Central Google Scholar

Download references

Acknowledgements

We are grateful to all members of the laboratory of S.L.M. We thank the WEHI core facilities of flow cytometry (S. Monard), histology (E. Tsui and E. Pan), genomics (S. Wilcox) and imaging (L. Whitehead) for excellent technical support. We thank S. Russo, E. Azzopardi and L. Wilkins for outstanding animal husbandry. This work was supported by grants from the Deutsche Forschungsgemeinschaft (SCHU 3126/3-1; to E.M.), the Spanish Ministry of Science, Innovation and Universities co-funded by the European Regional Development Fund/Agencia Estatal de Investigación (PID2021-125106OB-C31; to J.I.A.), the Australian National Health and Medical Research Council (1154325 and 1113577 to I.P.W.; 2003159 and 2003756 to S.L.M.), fellowships from the Victorian Endowment for Science Knowledge and Innovation (to S.L.M.), the HHMI-Wellcome International Research Scholarship (to S.L.M.) and the Sylvia and Charles Viertel Foundation (to S.L.M.). Figs. 6a and 7a and Extended Data Figs. 4a, 7b and 8b were created with BioRender.com.

Author information

Authors and Affiliations

Inflammation Division, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia
Elisabeth Mulazzani, Klara Kong, Paul J. Baker, Ian P. Wicks, Cynthia Louis & Seth L. Masters
Department of Medical Biology, University of Melbourne, Parkville, Victoria, Australia
Elisabeth Mulazzani, Nishika Ranathunga, Waruni Abeysekera, Alexandra L. Garnham, Paul J. Baker, Ian P. Wicks & Seth L. Masters
Department of Immunology, Hospital Clínic-IDIBAPS, Barcelona, Spain
Juan I. Aróstegui
Blood Cells and Blood Cancer Division, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia
Ashley P. Ng
Clinical Haematology Department, Royal Melbourne Hospital, Melbourne, Victoria, Australia
Ashley P. Ng
Peter MacCallum Cancer Centre, Parkville, Victoria, Australia
Ashley P. Ng
Division of Bioinformatics, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia
Nishika Ranathunga, Waruni Abeysekera & Alexandra L. Garnham
Immunology Research Unit, GlaxoSmithKline, Collegeville, PA, USA
Sze-Ling Ng & John D. Lich
Division of Immunology, The Walter and Eliza Hall Institute of Medical Research, Parkville, Victoria, Australia
Jacob T. Jackson
Department of Immunology and Pathology, Monash University, Clayton, Victoria, Australia
Margaret L. Hibbs
Rheumatology Unit, Royal Melbourne Hospital, Parkville, Victoria, Australia
Ian P. Wicks

Authors

Elisabeth Mulazzani
View author publications
You can also search for this author in PubMed Google Scholar
Klara Kong
View author publications
You can also search for this author in PubMed Google Scholar
Juan I. Aróstegui
View author publications
You can also search for this author in PubMed Google Scholar
Ashley P. Ng
View author publications
You can also search for this author in PubMed Google Scholar
Nishika Ranathunga
View author publications
You can also search for this author in PubMed Google Scholar
Waruni Abeysekera
View author publications
You can also search for this author in PubMed Google Scholar
Alexandra L. Garnham
View author publications
You can also search for this author in PubMed Google Scholar
Sze-Ling Ng
View author publications
You can also search for this author in PubMed Google Scholar
Paul J. Baker
View author publications
You can also search for this author in PubMed Google Scholar
Jacob T. Jackson
View author publications
You can also search for this author in PubMed Google Scholar
John D. Lich
View author publications
You can also search for this author in PubMed Google Scholar
Margaret L. Hibbs
View author publications
You can also search for this author in PubMed Google Scholar
Ian P. Wicks
View author publications
You can also search for this author in PubMed Google Scholar
Cynthia Louis
View author publications
You can also search for this author in PubMed Google Scholar
Seth L. Masters
View author publications
You can also search for this author in PubMed Google Scholar

Contributions

E.M. and S.L.M. designed the study. E.M., K.K., P.J.B., J.T.J., M.H., I.P.W., C.L. and S.L.M. performed and supervised the mouse experiments. J.I.A. collected clinical data and analyzed participant samples. A.N. performed the colony assay. S.-L.N. and J.N. designed the mouse model. N.K., A.L.G. and W.A. performed bioinformatic analysis. E.M. and S.L.M. wrote the manuscript with input from all authors.

Corresponding author

Correspondence to Seth L. Masters.

Ethics declarations

Competing interests

S.L.M. is a Scientific Advisor for Odyssey Therapeutics and NRG Therapeutics. S.-L.N. and J.N. are employees of GSK. I.P.W. has acted as a Scientific Advisor for CSL and received funding for research on G-CSF antagonism in inflammatory diseases. All other authors declare no competing interests.

Peer review

Peer review information

Nature Immunology thanks the anonymous reviewers for their contribution to the peer review of this work. Primary Handling Editor: N. Bernard, in collaboration with the Nature Immunology team. Peer reviewer reports are available.

Additional information

Publisher’s note Springer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.

Extended data

Extended Data Fig. 1 Further validation of PLCγ2^S707Y/+ mice.

a) Total bone mass of 3-week-old, male PLCγ2^S707Y/+ mutants compared to littermate controls (n = 3 mice per group). b) Whilst increased numbers of CD45⁺ across all organs are present, lymphoid immune cell infiltrates are sparse in paws, ears, tails, lung, gut, spleen. One representative IHC section (from three independent experiments; at 6 weeks of age) of CD45⁺, CD3⁺Ki67, B220⁺ stains is shown. c) PLCγ2^+/+ littermate controls lack immune cells infiltrates. One representative IHC section (from three independent experiments; at 6 weeks of age) of H&E, CD45⁺, MPO⁺, F4/80⁺, CD3⁺Ki67, B220⁺ is shown. d) Neutrophils (MPO⁺ in magenta) reside predominantly in the tissue rather than in the blood endothelial vessels (CD31⁺ with DAB) of 4-week-old PLCγ2^S707Y/+ mice as demonstrated by dual IHC (one representative from three independent experiments). e-g) ADVIA analyzer data of the blood reveal no difference with regards to platelets, red blood cells and reticulocytes in PLCγ2^S707Y/+ mutants compared to littermate controls (PLCγ2^+/+: n = 5; PLCγ2^S707Y/+: n = 6). Error bars represent mean ± SEM.

Source data

a) Cytokine assessment by a multiplex assay in plasma samples reveal increasing G-CSF levels following irradiation and reconstitution in PLCγ2^+/+ compared to G-CSF^−/− chimera (both groups at 11 weeks of age) receiving BM from PLCg2^S707Y/+ mice (n = 3 mice per group). b) Schematic illustration shows crosstalk between haematopoietic and non-haematopoietic cells in APLAID. Adapted from ‘Basophil-Mediated Skin Inflammation’, by BioRender.com (2023). Retrieved from https://app.biorender.com/biorender-templates.

Extended Data Table 1 Phenotypes/genotypes of published individuals with APLAID are summarized

Full size table

Extended Data Table 2 Phenotypes of different knockout strains