Abstract Background Because of repeated contact with airborne allergens, patients suffering from allergic asthma experience acute asthma attacks, characterized by shortness of breath, chest tightness, and coughing. The underlying immune response is highly complex and involves various immune cells. Chemokines play a pivotal role in the appropriate relocation of these diverse immune cells, ensuring their directed migration to the site of inflammation, their survival, and their effector functions. In the context of allergic asthma, the chemokine receptor CCR3 is crucially involved in T[H]2-mediated airway inflammation by recruiting eosinophils and other immune cells to the site of inflammation. However, more recent studies demonstrate its presence also on mast cells, macrophages, T cells, and dendritic cells. Objective We sought to investigate the role of CCR3 in different immune cell types during asthma pathogenesis. Methods Human peripheral blood cells collected from healthy controls and asthmatic individuals were analyzed for CCR3 expression. A murine model of asthma was used to compare wild-type and CCR3-deficient mice in the context of airway inflammation. Results In a human cohort of asthmatic patients, CCR3 mRNA expression was found induced in PBMCs and positively correlated with decreased lung function and blood eosinophilia. In a murine model of disease, CCR3 was found to be important for the establishment of eosinophilic inflammation. Moreover, CCR3-deficient mice showed impaired cytokine release, resulting in an innate-like mast cell and neutrophil-mediated lung inflammation and reduced T[H]2-orchestrated eosinophil-driven asthma. In the absence of CCR3, CD8 T cells underwent phenotypic changes, inhibiting the development of migratory effector memory CD8 T-cell subsets. Conclusions Taken together, this work demonstrates the functional involvement of CCR3 in both innate and adaptive immune cells in the lung during asthma pathogenesis. Key words: Chemokines, neutrophils, CD8, eosinophils, asthma __________________________________________________________________ Asthma bronchiale is a chronic, recurrent disease of the airways that results in a high burden of disease.[37]^1^,[38]^2 The 2 principal endotypes of asthma, type 2 (T2)-high or non-T2, are distinguished by the type of inflammation. Among these 2 endotypes, various phenotypes have been identified, including atopic asthma, nonatopic asthma, late-onset asthma, and obesity-related asthma.[39]^3 In T2-high asthma, the immune response is predominantly mediated by T[H]2 cells and innate lymphoid cells. These cells produce key T2 cytokines, including IL-4, IL-5, and IL-13, which play an essential role in driving eosinophilic inflammation and mucus hypersecretion, hallmark features of T2-high asthma. In contrast, non-T2 asthma lacks the elevated T2 immune activity and is instead characterized by alternative inflammatory profiles, such as neutrophilic or paucigranulocytic inflammation.[40]^4 During the disease pathogenesis of T2-high asthma, antigen sensitization, followed by a second allergen contact, causes IgE-mediated release of proinflammatory substances such as histamine, cytokines, proteases, and leukotrienes from mast cells, basophils, and eosinophils.[41]^5^,[42]^6 These mediators play an important role in driving and exacerbating the symptoms of asthma, such as bronchoconstriction, airway inflammation, and excessive mucus production. Chemokines are important mediators involved in cell migration, survival, and effector functions.[43]^7 The chemokine receptor CCR3 has been described primarily as a driver of eosinophil migration.[44]^8 However, more recent studies indicate that it plays a role in the function of additional cell types, including mast cells, macrophages, T cells, dendritic cells (DCs), as well as epithelial or airway smooth muscle cells. These findings and previous studies in asthmatic individuals suggest that CCR3 may also be involved in the pathogenesis of asthma.[45]9, [46]10, [47]11, [48]12, [49]13, [50]14, [51]15 CCR3 has a diverse ligand profile including chemokines such as eotaxin 1 to 3, RANTES, and monocyte chemoattractant protein 3 or 4. On release, these chemokines establish a chemokine gradient, guiding immune cells to the site of inflammation and initiating further effector functions.[52]^16 The objective of this study was to test the hypothesis that CCR3 plays a significant role in the pathogenesis of severe asthma. To achieve this, we conducted studies using human blood samples from both healthy and asthmatic individuals as well as a murine model of asthma to investigate the effects of CCR3 deficiency. In vivo, CCR3-deficient mice displayed a higher number of cDC1, facilitating more MHC-I cross-presentation to CD8 T cells in vitro. However, the recruitment of CD8 migratory effector memory T (T[MEM]) cells into the lung was impaired by CCR3 deficiency accompanied by an accumulation of CD8 tissue-resident memory (TRM) cells and exhausted T cells. Furthermore, CCR3 deficiency resulted in a diminished eosinophil response, shifting toward an innate-like asthma phenotype driven by neutrophil and mast cell inflammation, associated with elevated airway hyperresponsiveness (AHR). Methods A detailed description of extended methods is provided in this article’s Methods section in the Online Repository at [53]www.jaci-global.org. Human samples All human samples were collected during the human study AZCRA (“Investigation of the role of cytokines, chemokines and their receptors in the inflammatory process in asthma patients”), which was approved by the Ethics Committee of the Friedrich-Alexander-University Erlangen-Nürnberg (no. 20-315_4-B). Adult asthmatic and healthy control patients (aged 18-65 years) were recruited at the University Hospital Erlangen in collaboration with the Pneumology Team of the Department of Medicine 1 (Prof Dr M. F. Neurath and PD Dr S. Zirlik). The patients were informed about the study and gave their oral and written consent to participate. Exclusion criteria for both groups were severe and unstable asthma, other chronic lung diseases (eg, chronic obstructive pulmonary disease and cystic fibrosis), or other chronic diseases affecting the immune system. The clinical characteristics of the patients are provided in [54]Table E1 (in the Online Repository available at [55]www.jaci-global.org). PBMCs and PMNs were isolated by density centrifugation of whole blood samples using Biocoll (Bio&Sell, Feucht, Germany). The cells were then lysed with Qiazol lysis reagent for RNA isolation (Qiagen GmbH, Hilden, Germany). House dust mite–induced asthma model Murine experiments were performed according to the 3R principles and under the ethical approval of the local government of Unterfranken (Az 55.2-2532-2-633). The BALB/c CCR3 knockout (KO) mice (C.129S4-Ccr3tm1Cge/J) were a generous gift from Prof Dr Gerard Graham (Glasgow). House dust mite (HDM) extract freeze-dried from Dermatophagoides pteronyssinus (B82, Stallergenes Greer, Lenoir, NC) was purchased and reconstituted before use with PBS to obtain a stock solution of 2.5 mg/mL HDM. Mice were sensitized with 12.5 μg HDM intraperitoneally on days 0 and 7. For airway challenge, the mice were briefly narcotized with isoflurane, and 125 μg HDM extract in 25 μL PBS was applied on each nostril on days 14, 18, and 21.[56]^17 Lung function measurement in mice On day 22 of the HDM asthma protocol, mice underwent whole-body plethysmography with the Buxco WBP device (Data Sciences International, St Paul, Minn). After an adaption period, the mice were challenged with increasing doses of methacholine (0, 12.5, 25, and 50 mg/mL). To assess the lower airway resistance, an invasive measurement via the trachea was used. Therefore, animals were anesthetized with ketamine (150 mg/kg body weight) and medetomidine (0.25 mg/kg body weight) intraperitoneally. When they reached the surgical anesthesia state, they were injected intraperitoneally with rocuronium bromide (0.2 mg/kg body weight). Subsequently, the animals were tracheotomized and a blunt 18G cannula was inserted into the trachea and fixated with a thread. The cannula was then attached to a Flexivent FX1 piston respirator (Scireq, Montreal, Quebec, Canada). Each animal was challenged with increasing doses of methacholine (0, 5, 10, 25, and 50 mg/mL). Total cell isolation of murine lung and spleen tissue The lung cells were isolated by cutting the tissue into small pieces and digesting it enzymatically in collagenase/DNase solution (300 U/mL collagenase, Sigma-Aldrich Chemie GmbH, St Louis, Minn; 0.015% DNase, Roche Diagnostics GmbH, Mannheim, Germany) at 37°C for 45 minutes.[57]^18 The digested lung as well as the freshly isolated spleen were pushed through a 40-μm cell strainer and treated with ACK-lysis solution (NH[4]Cl 150 mM, KHCO[3] 10 mM, Na[2] EDTA 0.1 mM, filled with distilled water up to 1 L) before cell counting. In vitro coculture of CD8 T cells with antigen-presenting cells For coculture, CD8 T cells and antigen-presenting cells (APCs) were cultured in a 10:1 (500,000 T cells vs 50,000 APC) ratio. The A20 cell line was harvested from the cell culture flask and counted. To avoid excessive cell proliferation, we pretreated the A20 cells with mitomycin. Therefore, cells were incubated with 0.5 mg/mL mitomycin for 30 minutes at 37°C in an incubator. After washing them 3 times with PBS, cells were used for coculture. Isolated CD8 T cells were labeled with the cell tracer violet proliferation kit (Invitrogen, Thermo Fisher Scientific, Waltham, Mass) according to the manufacturer’s protocol. The labeled T cells were cultured together with the APCs in cell culture medium containing 50 μg/mL HDM as specific antigen. For the coculture of CD8 T cells with isolated CD11c^+ DCs, the same protocol was used without mitomycin treatment. After days 2 and 5, CD8 T cells were analyzed with a flow cytometer. Bulk RNA sequencing CD8^+ cells for RNA sequencing were isolated with fluorescence-activated cell sorting, which was done at the core unit for cell sorting and immunomonitoring at the Friedrich-Alexander-University Erlangen-Nürnberg. Splenic cells were stained with CD3 PE, CD8 BV421, CD11b APC Fire 750, and CD11c APC. The sorted cells were CD11b, CD11c double negative, and CD3, CD8 double positive. RNA for bulk RNA sequencing was isolated with the Qiagen RNeasy Mini Kit according to the manufacturer’s protocol. Bulk RNA sequencing was performed by the next-generation sequencing core facility of the Human Genetics Institute at Friedrich-Alexander-University Erlangen-Nürnberg. The quality of the RNA samples was checked with the Agilent bioanalyzer (Agilent Technologies, Inc, Santa Clara, Calif) and libraries were prepared with the Illumina stranded mRNA library prep kit (Illumina, Inc, San Diego, Calif) according to the manufacturer’s instructions with 12 cycles of PCR in the final amplification step. Libraries were pooled and sequenced as 150 + 150 bp paired-end reads on an Illumina Novaseq 6000 platform (Illumina, Inc). The data were processed by removing proximal optical duplicates with bbmap clumpify (v38.96), performing adapter and quality trimming using cutadapt (v3.3), and aligning the reads to the mouse genome (GRCh38) with STAR aligner (v2.7.10.a). Counts were generated with featureCounts (v2.0.1) if the exons were overlapping of the Ensembl Gene model (v102). Raw counts were then used for further analysis in the R environment (v4.3.0). Low counts were filtered by a minimum sum of 10 counts over all compared samples. Differential expression was calculated with DeSeq2 (v1.40.2) and annotated with org.Mm.eg.db (v3.17.0). Pathway enrichment analysis was performed with clusterProfiler (v4.8.3) and AnnotationDbi (v1.62.2) with a cutoff value of log[2] fold change greater than 0.8 or greater than −0.8. Heatmaps were created with ComplexHeatmap (v2.16.0). Principal-component analysis plots, dispersion plots, and volcano plots are shown in [58]Fig E1, A-F (in the Online Repository available at [59]www.jaci-global.org). Results Increased CCR3 expression in the peripheral blood inversely correlated with FEV[1]/forced vital capacity ratio in asthmatic patients Previous research has demonstrated the importance of CCR3 in eosinophil migration in the context of asthma.[60]^15^,[61]^19 Accordingly, we initiated an investigation into the expression of CCR3 in the peripheral blood of healthy controls and asthmatic individuals ([62]Fig 1, A). The clinical characteristics of the AZCRA cohort are provided in [63]Table E1. PBMCs, PMN cells, and serum were isolated from whole blood samples. The chemokine eotaxin 1 has been demonstrated to stimulate the migration of eosinophils from the bloodstream into the lung tissue.[64]^20 In the past, it was shown that individuals with asthma exhibited elevated levels of eotaxin 1 in the sputum.[65]^14 Similarly, we show that the asthmatic patients in our cohort had significantly increased levels of eotaxin 1 in serum ([66]Fig 1, B) and a significant increase in CCR3 gene expression in uncultured PBMCs ([67]Fig 1, C). However, no differential regulation of CCR3 mRNA expression in PMN cells was observed between healthy controls and asthmatic patients (see [68]Fig E2 in this article’s Online Repository at [69]www.jaci-global.org). In PBMCs, higher CCR3 expression was associated with decreased lung function in asthmatic patients, whereas no correlation was seen in the healthy control group ([70]Fig 1, D and E). A significant correlation between CCR3 expression in granulocytes and higher blood eosinophilia was found in asthmatic individuals but not in healthy controls ([71]Fig 1, F and G). Fig 1. [72]Fig 1 [73]Open in a new tab CCR3 drives inflammation and chemotaxis in asthmatic patients. A, Experimental design of the AZCRA cohort. B, Serum levels of eotaxin 1 measured by ELISA (n = 19, 34). C, Gene expression of CCR3/HPRT in PBMCs measured by qPCR (n = 15, 30). D, Correlation of CCR3/HPRT gene expression in PBMCs with FEV[1]/FVC (%) measured in healthy controls (n = 15). E, Correlation of CCR3/HPRT gene expression in PBMCs with FEV[1]/FVC (%) measured in asthmatic patients (n = 30). F, Correlation of CCR3/HPRT gene expression in granulocytes with blood eosinophils (%) measured in healthy controls (n = 18). G, Correlation of CCR3/HPRT gene expression in granulocytes with blood eosinophils (%) measured in asthmatic patients (n = 30). FVC, Forced vital capacity; HPRT, Hypoxanthine phosphoribosyltransferase; qPCR, quantitative PCR. Data are expressed as mean ± SEM. Significance was calculated by using the Welch t test (Fig 1, B), the Mann-Whitney U test (Fig 1, C), or the nonparametric Spearman correlation (Fig 1, D-G). ∗P < .05; ∗∗P < .01. CCR3 deficiency results in increased disease burden The human data indicated a potential involvement of CCR3 in the development of asthma, particularly in mononuclear cells. Therefore, we started investigating the role of CCR3 in an HDM-induced murine asthma model by comparing CCR3-deficient mice and wild-type (WT) littermates ([74]Fig 2, A). Following the HDM challenge, an increase in CCR3 mRNA expression was observed in the lungs of WT mice, whereas eotaxin levels in serum were selectively induced in CCR3 KO mice only (see [75]Fig E3, A and B, in this article’s Online Repository at [76]www.jaci-global.org). Moreover, CCR3-deficient mice exhibited markedly elevated enhanced pause (Penh) ([77]Fig 2, B) and increased respiratory resistance in comparison with WT mice ([78]Fig 2, C). Penh indicates difficulty of the air to exit during expiration. It is a parameter obtained during unrestrained whole-body plethysmography and serves as an indicator for the obstruction of the airways.[79]^21 Both Penh and respiratory resistance indicated reduced lung function in CCR3-deficient mice. In hematoxylin and eosin–stained histological sections, HDM-challenged CCR3-deficient mice showed increased inflammation in the perivascular and peribronchial tissue as compared with the HDM-challenged WT mice ([80]Fig 2, D and E). Furthermore, the bronchoalveolar lavage (BAL) of WT mice predominantly constituted eosinophils, whereas CCR3 KO mice presented with a mixed population of neutrophils, lymphocytes, and mononuclear cells ([81]Fig E3, C) as well as elevated mast cell numbers in the lung ([82]Fig 2, F). Serum IgE levels were found to be lower in CCR3-deficient mice as compared with WT mice ([83]Fig 2, G). However, CCR3 deficiency in the context of asthma resulted in increased numbers of CD44^+CD4 effector T cells ([84]Fig E3, D) with elevated T[H]2 polarization, as evidenced by higher prevalence of GATA-3^+ CD4 T cells in the lung as compared with WT mice ([85]Fig 2, H). Fig 2. [86]Fig 2 [87]Open in a new tab CCR3 deficiency in mice worsens asthmatic burden. A, Experimental design of the HDM-induced asthma model. B, Noninvasive WBP on day 22 (n = 8, 9, 7, 10). C, Invasive measurement of AHR on day 23 (n = 7, 9, 7, 9). D, Representative pictures of H&E-stained lung sections. E, Inflammation score of H&E-stained lung sections (n = 8, 8, 9, 10). F, Flow-cytometric analysis of c-Kit^+ FcεRI^+ mast cells in the lung (n = 8, 10). G, Serum levels of IgE measured by ELISA (n = 8, 10, 8, 9). H, Flow-cytometric analysis of GATA-3^+ T[H]2 cells in the lung (n = 12, 11, 12, 15). H&E, Hematoxylin and eosin; Rrs, respiratory resistance; WBP, whole-body plethysmography. Data are expressed as mean ± SEM. Significance was calculated by using the Mann-Whitney U test (Fig 2, F) or the ordinary 2-way ANOVA with post hoc Šidák multiple comparisons test. ∗P < .05; ∗∗P < .01; ∗∗∗P < .005; ∗∗∗∗P < .001. Impaired cytokine release and CCR3 deficiency diminish eosinophilic inflammation Typically, T[H]2 polarization results in the recruitment of eosinophils from the blood.[88]^1 However, analysis by flow cytometry revealed that CCR3 KO mice had a lower number of eosinophils in the lungs in comparison with WT mice ([89]Fig 3, A). Lung eosinophils can be classified into 2 distinct subsets, resident and inflammatory, on the basis of their expression of CD62L and CD101.[90]^22 Both subsets show different properties and have markedly different gene expression signatures. Inflammatory eosinophils are predominantly localized in the peribronchial area, where they actively contribute to lung inflammation. In contrast, resident eosinophils are not dependent on IL-5 and reside in the parenchymal tissue during both homeostasis and inflammation.[91]^22^,[92]^23 In humans, an increase in the number of inflammatory eosinophils has been demonstrated to correlate with disease severity.[93]^24 In HDM-challenged mice, we found that the number of resident eosinophils was reduced in the lung as well as BAL (see [94]Fig E4, A, in this article’s Online Repository at [95]www.jaci-global.org; see also [96]Fig 3, B). The reduction was less pronounced in CCR3 KO mice as compared with WT mice. Moreover, the number of inflammatory eosinophils was significantly reduced in the lung and BAL of CCR3 KO mice ([97]Fig 3, C; see also [98]Fig E4, B). Fig 3. [99]Fig 3 [100]Open in a new tab CCR3-deficient mice show abundant eosinophil response and impaired cytokine release. A, Flow-cytometric analysis of Siglec-F^+ GR1^− lung eosinophils (n = 7, 8, 9, 10). B, Flow-cytometric analysis of CD101^−CD62L^+ resident eosinophils gated on Siglec-F^+ CD45^+ nonlymphocytes in the lung (n = 8, 7, 9, 10). C, Flow-cytometric analysis of CD101^+CD62L^− inflammatory eosinophils gated on Siglec-F^+ CD45^+ nonlymphocytes in the lung (n = 8, 7, 9, 10). D, ELISA analysis of IL-5 production in anti-CD3 anti-CD28–stimulated lung cells cultured for 24 hours. Iono groups were stimulated with ionomycin 6 hours before harvesting the cells (n = 8, 9, 7, 9, 8, 9, 8, 9). E, Schematic illustration of cytokine release in WT, CCR3-deficient, and ionomycin-stimulated CCR3-deficient cells. Representative dot plots are shown. NS, Not significant. Data are expressed as mean ± SEM. Significance was calculated by the ordinary 2-way ANOVA with post hoc Šidák multiple comparisons test. ∗∗P < .01; ∗∗∗∗P < .001. The secretion of the cytokine IL-5 by T[H]2 cells is important for eosinophil survival.[101]^25 Consequently, the levels of IL-5 were quantified in cell culture supernatants of lung cells stimulated with anti-CD3 anti-CD28. However, despite a higher number of T[H]2 cells, IL-5 production was significantly lower in HDM-challenged CCR3-deficient mice as compared with WT mice ([102]Fig 3, D). CCR3 belongs to the family of G protein–coupled receptors. On activation, G protein–coupled receptors induce a calcium influx from the endoplasmic reticulum into the cytosol.[103]^26 CCR3-deficient mice lack the activation of CCR3 to induce this calcium increase, which is essential for the release of cytokines. We found that increased calcium influx due to ionomycin treatment was capable of restoring IL-5 release in CCR3-deficient lung cells ([104]Fig 3, D and E). CCR3 deficiency shifts lung inflammation toward a neutrophil-driven phenotype In our model, CCR3-deficient mice showed more pronounced inflammation, dominated by the presence of neutrophils, mast cells, and T cells. Consistently, we found an increased number of neutrophils in the lung and BAL ([105]Fig 4, A; see also [106]Fig E4, C). In addition, the production of neutrophil-attracting TNF-α was higher in CCR3-deficient mice, with potential sources including T cells, macrophages, and mast cells ([107]Fig 4, B). Simultaneously, the presence of γδ T cells in the lungs was augmented in CCR3-deficient mice ([108]Fig 4, C). In these mice, the specific Vγ4 subset was significantly more abundant ([109]Fig 4, D). γδ T cells are producers of IL-17, a cytokine that promotes neutrophil recruitment, and have also been described as capable of performing MHC-I cross-presentation to CD8 T cells, which enables them to induce neutrophilic inflammation.[110]^27 Fig 4. [111]Fig 4 [112]Open in a new tab Absence of CCR3 induces neutrophil-driven inflammation. A, Flow-cytometric analysis of GR1^+ Siglec-F^− CD3^− nonlymphocytes in the lung (n = 7, 9, 8, 10). B, ELISA analysis of TNF-α in the supernatant of anti-CD3 anti-CD28–stimulated total lung cells cultured for 48 hours (n = 4, 4, 5, 4). C, Flow-cytometric analysis of total γδ TCR^+ CD3^+ T cells in the lungs of WT and CCR3 KO mice (n = 8, 9, 9, 7). D, Flow-cytometric analysis of Vγ4^+ γδ TCR^+ CD3^+ T cells in the lungs of WT and CCR3 KO mice (n = 8, 9, 9, 7). Representative dot plots are shown. Data are expressed as mean ± SEM. Significance was calculated by the ordinary 2-way ANOVA with post hoc Šidák multiple comparisons test. ∗P < .05; ∗∗P < .01; ∗∗∗P < .005; ∗∗∗∗P < .001. CCR3 deficiency facilitates enhanced MHC-I antigen presentation and induced CD8 T-cell proliferation in vitro CCR3 KO mice displayed elevated T[H]2 polarization and a more severe inflammation as compared with WT mice. Therefore, we proceeded to analyze the antigen-presenting DC populations in the lungs and found that the total number of DCs as well as cDC2 were unaltered (see [113]Fig E5, A and B, in this article’s Online Repository at [114]www.jaci-global.org). However, we discovered that CCR3 KO mice had an increased number of the MHC-I–expressing cDC1 population ([115]Fig 5, A). We thus analyzed the antigen presentation of DCs toward CD8 T cells in CCR3 deficiency. We found that CD8 T cells, isolated from HDM-challenged CCR3-deficient mice, displayed significantly increased cell proliferation after in vitro coculture with CCR3 KO DCs ([116]Fig 5, B). When culturing WT CD8 T cells with CCR3-deficient DCs, they also showed an induced cell proliferation, whereas coculture of CCR3-deficient CD8 T cells with WT DCs resulted in reduced proliferation ([117]Fig 5, C). However, CCR3-deficient T cells alone also showed a higher cell proliferation after culture with anti-CD3 and anti-CD28 and displayed higher proliferation after coculture with the antigen-presenting cell line A20 ([118]Fig E5, C and D). These findings indicated that CCR3 deficiency might induce cDC1 antigen cross-presentation. Consequently, we conducted a flow-cytometric analysis to determine the number of CD8 T cells in CCR3-deficient mice. WT CD8 T cells exhibited augmented CCR3 expression in the lung following HDM challenge ([119]Fig 5, D). In contrast to higher cell proliferation in vitro, CCR3 KO mice had a significantly lower number of CD8 T cells as compared with WT mice ([120]Fig 5, E). The reduced number of CD8 T cells could be attributed to a reduced CD8 T-cell migration toward a chemokine gradient. Accordingly, we investigated the CD8 T-cell migration toward the CCR3 ligand RANTES in a transwell experiment. However, migration of CD8 T cells toward RANTES was not impaired by CCR3 deficiency ([121]Fig E5, E). We thus concluded that the reduction in the number of CD8 T cells in vivo is caused by other factors. Fig 5. [122]Fig 5 [123]Open in a new tab CCR3 deficiency induces MHC-I cross-presentation of cDC1 toward CD8 T cells. A, Flow-cytometric analysis of XCR1^+, CD11c^+, MHC-II^+, CD64^−, F4/80^− cDC1 in the lung (n = 4, 6, 7, 8). B, Percentage of splenic CD8 T cells in generation G3 after 2 days of coculture with the same CD11c^+ cells evaluated by flow cytometry (n = 4, 4, 4, 5). C, Percentage of splenic CD8 T cells in generation G3 after coculture with the respective different phenotype CD11c^+ DCs evaluated by flow cytometry (n = 3, 3, 3, 3). D, Flow-cytometric analysis of CCR3^+CD8^+CD3^+ T cells in the lung (n = 8, 9). E, Flow-cytometric analysis of total CD8^+CD3^+ T cells in the lung of WT and CCR3 KO mice (n = 8, 8, 9, 10). Representative dot plots or histograms are shown. FMO, Fluorescence Minus One. Data are expressed as mean ± SEM. Significance was calculated by unpaired t test (Fig 5, D) or the ordinary 2-way ANOVA with post hoc Šidák multiple comparisons test. ∗P < .05; ∗∗P < .01; ∗∗∗P < .005; ∗∗∗∗P < .001. CD8 T cells display an altered phenotype in CCR3-deficient mice To gain further insight into the CD8 T-cell subsets, we performed bulk RNA sequencing of sorted CD8 T cells. The gene signature revealed a downregulation of the transcription factor Eomes in CD8 T cells of CCR3-deficient mice. Furthermore, distinct chemokine receptors such as Ccr2, Ccr5, and Ccr10 were upregulated in naive and HDM-challenged CCR3 KO mice ([124]Fig 6, A and B). Naive CD8 T cells exhibited an induced gene signature indicative of exhaustion, as evidenced by elevated expression of Tox2 and Pdcd1. Flow cytometry confirmed increased PD-1 protein expression also in the lung of CCR3-deficient mice ([125]Fig E6, A). Consistently, lung CD8 T cells expressed only low levels of cytotoxic granule proteins perforin and granzyme B ([126]Fig E6, B). The gene expression of different integrins such as Itgae and Itga4 was induced in CCR3-deficient CD8 T cells. Flow-cytometric analysis of CCR3-deficient lung CD8 T cells also showed increased expression of integrin heterodimer VLA-4, comprising CD29 and CD49d ([127]Fig E6, C). Fig 6. [128]Fig 6 [129]Open in a new tab CD8 T-cell signature is altered in CCR3 deficiency, resulting in enhanced TRM phenotype. A, Heatmap of selected, significantly differentially expressed genes between splenic CD8 T cells isolated from naive CCR3 KO and WT mice (n = 4, 4). B, Heatmap of selected, significantly differentially expressed genes between splenic CD8 T cells isolated from HDM-treated CCR3 KO and WT mice (n = 3, 4). C, Flow-cytometric analysis of CX3CR1^+CD27^−CD8^+CD3^+ T cells in the lymph nodes of WT and CCR3 KO mice (n = 4, 4, 5, 4). D, Flow-cytometric analysis of CD103^+CD8^+CD3^+ T cells in the lung of WT and CCR3 KO mice (n = 8, 8, 9, 10). E, ELISA analysis of IL-4 production in the supernatant of anti-CD3 anti-CD28–stimulated lung cell cultured for 48 hours (n = 4, 6, 5, 4). F, Gene expression of Eomes/HPRT mRNA in the total lung, measured by quantitative PCR (n = 7, 8, 7, 9). Representative dot plots are shown. HPRT, Hypoxanthine phosphoribosyltransferase. Data are expressed as mean ± SEM. Significance was calculated by the ordinary 2-way ANOVA with post hoc Šidák multiple comparisons test. ∗P < .05; ∗∗P < .01; ∗∗∗P < .005. In CD8 T cells isolated from HDM-challenged mice, significant upregulation was observed for Il2ra, Socs2, and Itgae ([130]Fig 6, B). However, the expression of Il4ra and Eomes was found to be decreased. As was previously observed in naive animals, a number of chemokine receptors were upregulated in CCR3 KO mice. Furthermore, pathway enrichment analysis confirmed a pronounced chemokine receptor expression and the regulation of protein localization to the plasma membrane ([131]Fig E7, A and B, in this article’s Online Repository at [132]www.jaci-global.org). The gene expression profile indicated a reduction in Eomes activity, which could lead to alterations in memory formation. Indeed, CCR3-deficient mice had reduced numbers of CD8 T[MEM]cells in the lymph nodes in comparison with WT mice ([133]Fig 6, C). Furthermore, it was observed that HDM-challenged mice lacking CCR3 exhibited a markedly elevated population of CD103^+CD8 TRM cells ([134]Fig 6, D). It has been reported that Eomes expression is induced and maintained by IL-4 signaling.[135]^28 Similar to IL-5 secretion, IL-4 release in CCR3-deficient mice was significantly reduced ([136]Fig 6, E). Consistent with RNA sequencing, Eomes expression in the lungs of CCR3 KO mice was strongly reduced ([137]Fig 6, F). Collectively, these results depict a phenotypic change of CCR3-deficient CD8 T cells toward a tissue-resident signature while T[MEM] cells are impaired. Discussion The objective of this study was to investigate the role of CCR3 in the pathogenesis of asthma. The results on transcriptional changes in CCR3-deficient CD8 T cells add to previous reports that focused on the effect of CCR3 deficiency on eosinophils.[138]^19^,[139]^29 Our human data indicate higher expression of CCR3 mRNA in PBMCs isolated from asthmatic patients, which was associated with symptom severity. These results are in line with a previous report on higher expression of CCR3 in peripheral CD4 and CD8 T cells of atopic individuals as compared with healthy controls.[140]^30 For further investigation, we used an HDM-induced asthma model to analyze WT and CCR3-deficient mice. In accordance with the findings of Humbles et al,[141]^19 deficiency of CCR3 resulted in augmented AHR and more pronounced inflammation when compared with WT mice. Despite the presence of a greater number of T[H]2 cells, CCR3-deficient mice failed to establish a T[H]2 cytokine response because of the impairment of IL-5 and IL-4 release. Cytokines are transported via the endoplasmic reticulum and stored in specific sites or released immediately. They can be released constitutively as a result of induced transcription or following receptor signaling.[142]^31 Chemokine receptors belong to the cytokine release–triggering receptors. It was therefore hypothesized that CCR3-deficient mice were lacking the stimulating signal to release preformed cytokines via CCR3-triggered calcium influx.[143]^32 In fact, stimulation with a calcium ionophore in vitro was able to restore IL-5 production to levels comparable with those seen in WT mice. Confirming the results of other studies, we found a defect in eosinophil numbers in CCR3-deficient mice when challenged with HDM.[144]^19^,[145]^29^,[146]^33 Moreover, we reported reduced inflammatory eosinophil numbers in the lungs and BAL of CCR3 KO mice as compared with WT animals.[147]^22 CCR3 KO mice, lacking CCR3-induced eosinophil chemotaxis, also failed to release IL-5, thus not creating an environment conducive to eosinophil accumulation. However, we observed a transition from an eosinophil-dominated inflammation toward a neutrophil-driven phenotype in CCR3-deficient mice. CCR3-deficient mice had elevated neutrophil and mast cell numbers. The increased mast cell populations may serve as potential sources for TNF-α production, which could potentially result in the recruitment of neutrophils into the lung.[148]^34 According to the literature, T[H]17 immune responses have been described to induce neutrophil migration into the lung.[149]^35 Consistently, an induction in γδ T-cell numbers, which are known to exert T[H]17-like immune responses, driving neutrophilic inflammation via the production of IL-17, was seen in CCR3 KO mice.[150]^27 Neutrophils and mast cells produce proinflammatory mediators that contribute substantially to asthma pathogenesis.[151]^36^,[152]^37 Therefore, the elevated numbers of mast cells and neutrophils may collectively contribute to the enhanced AHR observed in CCR3-deficient mice. Our data show increased numbers of MHC-I cross-presenting cDC1 in CCR3-deficient mice. Thus, we analyzed the antigen presentation between cDC1 and CD8 T cells. Consistently, in vitro, we found increased cell proliferation of CD8 T cells cocultured with isolated DCs. Recently, CCR3 was reported to be expressed on cDC1 in patients with active Crohn disease.[153]^38 However, the role of CCR3 in cDC1 remains unclear and needs further clarification. In our asthma model, CCR3 expression was significantly induced in CD8 T cells. It was therefore anticipated that there would be functional differences between WT and CCR3-deficient CD8 T cells. Indeed, RNA sequencing revealed changes in CD8 T-cell signature in CCR3-deficient mice. In CCR3-deficient CD8 T cells, IL-4 production and receptor expression were reduced. IL-4Rα signaling has been demonstrated to be crucial for the sustenance of the CD8 population. Conversely, IL-4 deficiency has been linked to diminished expression of Eomes, a reduction in the number of CD8 memory T cells, and an increase in TNF-α production.[154]^28^,[155]^39 Moreover, missing IL-4 signaling potentially causes lower serum IgE levels in CCR3-deficient mice as IL-4 mediates B-cell maturation. CD8 T cells from CCR3-deficient mice showed an exhausted phenotype, characterized by the expression of PD-1, Tox-2, and CTLA-4 (cytotoxic T lymphocyte–associated antigen 4). Accordingly, they also displayed less cytotoxicity as seen by reduced perforin and granzyme B production and downregulation of granzyme A. Higher T-cell exhaustion might account for reduced number of CD8 T cells in the lungs of CCR3-deficient mice. T[MEM] phenotype of CD8 T cells derived from CCR3 KO mice was reduced in the lymph nodes. It was shown that CX3CR1[int] T[MEM] can home to the lymph nodes but are also important in the surveying of peripheral tissues.[156]^40 Moreover, current literature has provided evidence that CCR3 gene expression is induced in lung T[MEM] cells.[157]^41 Our results suggest that CCR3 plays a role in the development and migration of these cells. However, CCR3 KO mice had a higher number of CD8 TRM cells in the lung. Consistent with the RNA-sequencing gene signature, TRM cells typically express only low levels of Eomes.[158]^42 In addition, the limited production of the cytokine IL-4, which drives the expression of Eomes in CD8 T cells, further impairs effector memory development.[159]^39 Further experiments in CD8 conditional KO cells should be carried out to explore this finding in more detail. The existing literature provides evidence that CCR3 plays a proinflammatory role by recruiting eosinophils. In human cohorts, our own findings and those of others indicate a correlation between CCR3 expression and disease severity.[160]^14 However, our murine data also provide evidence on a dual role of CCR3 with an important role of CCR3 in the maintenance of CD8 T-cell subsets. To corroborate these findings in humans, further investigations on protein levels in PBMCs and specific cell types are required. Most human studies have concentrated on eosinophils and T[H]2 cells. In contrast, our murine model permits the investigation of a complete deletion of CCR3 in various cell types. Given the broad expression of CCR3 on various cell types, it is plausible that it may possess disparate functionalities in different cell types, potentially leading to discrepancies between the human data and our murine model. This study demonstrated the effect of CCR3 deficiency on the maturation of CD8 effector function. CCR3-deficient CD8 T cells showed induced proliferation when cocultured with CD11c cells. They were characterized by a different gene expression pattern as compared with WT CD8 T cells. Moreover, the absence of CCR3 signaling caused a defect in cytokine release, which was crucial for the establishment of a robust T[H]2 response. Eosinophilic inflammation was reduced, whereas CCR3 deficiency gave rise to neutrophil- and mast cell–dominated inflammation, which resulted in severe AHR. Key messages. * • Induced CCR3 mRNA expression in PBMCs is associated with reduced FEV[1]/forced vital capacity ratio values in asthmatic patients. * • CCR3 KO mice exhibited neutrophilic inflammation and increased mast cell infiltration in the lung. * • CCR3-deficient mice showed altered CD8 T-cell response characterized by increased CD8 TRM population. Data availability: All relevant data are provided in the article and the Online Repository at [161]www.jaci-global.org. Additional source data are available on request from the corresponding author. Disclosure statement This work was funded by a Deutsche Forschungsgemeinschaft grant (project no.: 517590146) awarded to S.F. in Erlangen. Disclosure of potential conflict of interest: The authors declare that they have no relevant conflicts of interest. Acknowledgments