Macrophage-Specific, Mafb-Deficient Mice Showed Delayed Skin Wound Healing

Macrophages play essential roles throughout the wound repair process. Nevertheless, mechanisms regulating the process are poorly understood. MAFB is specifically expressed in the macrophages in hematopoietic tissue and is vital to homeostatic function. Comparison of the skin wound repair rates in macrophage-specific, MAFB-deficient mice (Mafbf/f::LysM-Cre) and control mice (Mafbf/f) showed that wound healing was significantly delayed in the former. For wounded GFP knock-in mice with GFP inserts in the Mafb locus, flow cytometry revealed that their GFP-positive cells expressed macrophage markers. Thus, macrophages express Mafb at wound sites. Immunohistochemical (IHC) staining, proteome analysis, and RT-qPCR of the wound tissue showed relative downregulation of Arg1, Ccl12, and Ccl2 in Mafbf/f::LysM-Cre mice. The aforementioned genes were also downregulated in the bone marrow-derived, M2-type macrophages of Mafbf/f::LysM-Cre mice. Published single-cell RNA-Seq analyses showed that Arg1, Ccl2, Ccl12, and Il-10 were expressed in distinct populations of MAFB-expressing cells. Hence, the MAFB-expressing macrophage population is heterogeneous. MAFB plays the vital role of regulating multiple genes implicated in wound healing, which suggests that MAFB is a potential therapeutic target in wound healing.


Introduction
V-maf musculoaponeurotic fibrosarcoma oncogene family, protein B (MAFB) is a member of the Maf transcription factor (TF) family that includes basic-region leucine zipper-type TFs, which bind the Maf recognition element (MARE) by dimerization [1]. MAFB was reported as an oncogene involved in inducing multiple myeloma [2], and Human Protein Atlas (HPA) data indicate that the expression of MAFB is related to the growth of endometrial cancer [3].
MAFB is widely expressed in several tissues and is specifically expressed in the macrophages of the hematopoietic tissue [4]. In particular, HPA data show that MAFB is highly expressed in Kupffer cells in humans [5]. In vitro analyses demonstrated that MAFB is specifically expressed in anti-inflammatory IL-4-or IL-10-induced, bone marrow-derived M2-type macrophages [6]. A previous study showed that MAFB regulates the C1q gene, the first protein in the classical pathway, to promote efferocytosis and prevents autoimmune diseases in tissue-resident macrophages [7]. MAFB is also expressed in macrophages associated with various disease conditions. In ischemic stroke, MAFB may regulate the expression of mouse scavenger receptor 1 (MSR1) and macrophage receptor with collagenous structure (MARCO), thereby reducing the levels of damage-associated molecular patterns (DAMPs) [8]. In atherosclerosis, MAFB prevents the apoptosis of macrophages by regulating the apoptosis inhibitor of macrophage (AIM) gene [9]. MAFB also inhibits the self-renewal function of terminally differentiated macrophages [10]. Therefore, MAFB can regulate macrophage function and maintain homeostasis in disease. With these functions, it is possible that MAFB also relates to other diseases and abnormal conditions, such as wound healing.
Repair and homeostasis in response to skin injury are essential. Rapid wound repair is vital in the event of severe trauma. In previous studies, macrophages were removed during wound healing and were empirically demonstrated to be essential for this process [11,12]. Lucas et al. reported that macrophage removal during the inflammatory phase decreased angiogenesis and delayed the formation of the granulation tissue [13]. However, macrophage removal during the proliferative phase resulted in severe bleeding into the wound tissue [13]. These findings suggest that macrophages regulate different cells at each stage of tissue repair and that a lack of cellular response at any stage can delay repair and lead to chronic wounds. Therefore, macrophage function must be tightly regulated. Nevertheless, the mechanisms determining which macrophage subsets are functional at each phase of wound repair remain to be elucidated. Macrophages express cytokines and chemokines that play important roles in wound healing. CCL2 promotes re-epithelialization by recruiting macrophages [14]. CCL12 suppresses inflammation and induces fibrosis [15]. IL-10 suppresses excessive fibrosis during remodeling [16]. Arg1 is a well-known M2 macrophage marker that induces fibrosis during wound healing [17]. However, transcriptional regulation of the aforementioned factors is yet to be clarified. Furthermore, the TLR2 −/− mice delayed wound healing [18], and HPA indicates that MAFB is coexpressed with TLR2 in humans [5]. However, the expression of these in mice wounds is unknown.
In the present study, mice with specific Mafb knockouts in certain macrophage lineages (Mafb f/f ::LysM-Cre) were used to investigate the roles and functions of MAFB in wound healing. We discovered that MAFB was expressed in the granulation tissue and that wound healing was delayed in Mafb f/f ::LysM-Cre mice. Moreover, Ccl12, Arg1, and Ccl2 were downregulated in the wound sites by day 3 after skin injury in the Mafb f/f ::LysM-Cre mice. Hence, macrophage MAFB induces genes associated with wound repair and is crucial for proper wound healing.

MAFB Is Required for Proper Wound Healing
We wounded Mafb f/f and Mafb f/f ::LysM-Cre mice and measured the wound areas on days 0, 3, 5, 7, and 10 to confirm whether MAFB is related to wound healing ( Figure 1A,B). The average wound area in Mafb f/f ::LysM-Cre mice was 1.3, whereas it was 1 in Mafb f/f mice (p = 0.009) on day 3. On day 5, the area in Mafb f/f ::LysM-Cre was 1.4, whereas it was 1 in Mafb f/f mice (p = 0.024). By day 10, several Mafb f/f mice presented with closed wounds, whereas only two Mafb f/f ::LysM-Cre mice demonstrated wound closure. Hence, wound healing was significantly impaired in the Mafb f/f ::LysM-Cre mice. We used hematoxylin and eosin (H&E) staining to examine the histological differences between Mafb f/f and Mafb f/f ::LysM-Cre wounds and observed their hallmarks on days 3 and 5 ( Figure 1C,D). We found that the areas of the dermis in the H&E sections were significantly larger in the Mafb f/f ::LysM-Cre mice than they were in the Mafb f/f mice on day 3 (p = 0.023; Figure 1D). We used Masson's trichrome staining to measure the collagen area in the granulation tissue and found that it was significantly decreased in the Mafb f/f ::LysM-Cre mice relative to that in the Mafb f/f mice on day 5 ( Figure 1E). Therefore, MAFB was required for prompt and timely wound healing and was implicated in collagen accumulation.   Graph showing the % collagen area per granulation tissue area. Data are means ± S.E.M; * p < 0.05, ** p < 0.01, Student's t-test. Each dot represents data for a single mouse.

MAFB Expressed on Ly6C + Macrophages in Granulation Tissue
Immunostaining was performed on the day 3 wounds of Mafb +/GFP mice to confirm MAFB expression. GFP expression was higher in the granulation tissue (Figure 2A,B) than in the dermis ( Figure 2C). To confirm whether the GFP-expressing cells were macrophages, we performed FACS analysis on the day 3 wounds on Mafb +/GFP mice. To this end, we used the myeloid marker antibody CD11b and the macrophage marker antibodies CD204, Mac2, and F4/80 ( Figure 2D) and observed the cell populations coexpressing GFP with all the macrophage markers. We performed FACS analysis on the day 3 wounds of Mafb +/GFP mice to identify the GFP-expressing cells. For this purpose, we used antibodies against CD11b, Gr-1, and Ly6C ( Figure 2E). The green dots indicate GFP + cells. Whereas the Gr-1 − , CD11b + , Ly6C + monocyte-derived macrophages expressed GFP, the CD11b + , neutrophil marker Gr-1 + cells did not. Hence, MAFB was expressed in the granulation tissue and, more precisely, in the macrophages around the wounds.

MAFB Expression in Macrophages Is Related to Their Recruitment in Wounds
FACS analysis was performed on day 0, 3, 5, and 7 wounds to compare Mafb f/f and Mafb f/f ::LysM-Cre mice in terms of the numbers and dynamics of the cells in their wounds during healing. The neutrophils (CD11b + , Gr-1 + cells) and monocytes/macrophages (CD11b + , Gr-1 − cells) were enumerated and are displayed in Supplementary Figure S1. Gating strategies are shown in Supplementary Figure S1A. The numbers of whole monocytes/macrophages were low under normal conditions (day 0) but gradually increased until day 7 after injury (Supplementary Figure S1B,C). The monocyte/macrophage population was divided into the CD11b high and CD11b low subpopulations (Supplementary Figure S1B). The latter were further subdivided into the CD11b high , Ly6C − , CD11b high , Ly6C + , and CD11b low , Ly6C + cell subpopulations ( Figure 3A). Relative to day 0, the cell numbers in all populations increased by day 3. However, only the CD11b high, Ly6C + cells peaked at day 3, and they decreased thereafter. In contrast, the numbers of the CD11b high , Ly6C − cells and CD11b low , Ly6C + cells continued to increase after day 3 and reached their peak by day 7. The foregoing experiments demonstrated cell dynamics and numbers during wound healing. By day 5, the CD11b high , Ly6C + cells were significantly lower in the Mafb f/f ::LysM-Cre mice than in the Mafb f/f mice. Thus, MAFB might recruit monocytes/macrophages.

Mafb Deficiency Is Associated with CCL12, CCL2, and Arg1 Downergulation in Wounds
The effects of Mafb deletion on the number of cells were evident by day 5. Therefore, we used kits to measure the proinflammatory and anti-inflammatory cytokines and angiogenesis-related proteins in the wound tissues on day 3 (Supplementary Figure S2). Figure 4A is an enlarged photograph of the membrane. CCL12 and CXCL10 were slightly expressed in the Mafb f/f mice but were not expressed at all in the Mafb f/f ::LysM-Cre mice. The IL-10 and CCL2 expression levels were lower in the Mafb f/f ::LysM-Cre mice than in the Mafb f/f mice. Several differentially expressed genes detected between the Mafb f/f mice and the Mafb f/f ::LysM-Cre mice in the proteome analysis were selected and their mRNA levels in day 3 wound tissue were quantitated using RT-qPCR ( Figure 4B). Ccl12 was significantly decreased and Ccl2 tended to decrease in the Mafb f/f ::LysM-Cre mice. However, the expression levels of Cxcl10, Il-10, and the other genes did not differ between the Mafb f/f and Mafb f/f ::LysM-Cre mice. Moreover, the expression levels of the proinflammatory genes, Il-12b and Nos2, did not differ between the Mafb f/f and Mafb f/f ::LysM-Cre mice. Here, wound tissue samples also contain various cells other than macrophages. Therefore, gene expression in other cells may mask differences in the gene expression of macrophages. For this reason, we conducted RT-qPCR on bone marrow-derived M2 macrophages induced with M-CSF, IL-4, and IL-13 in Mafb f/f and Mafb f/f ::LysM-Cre mice. Ccl12 and Ccl2 were downregulated both in the Mafb f/f ::LysM-Cre mice and the wound tissue. In contrast, the Il-10, Arg1, and c-Maf expression levels did not differ among wound tissues but significantly differed in M2 macrophages in vitro ( Figure 4C). It has been reported that c-Maf and Mafb are mutually compensatory. However, this compensation was not apparent under the M2 condition. Il-10 is the direct target of c-Maf and was consistently downregulated in Mafb f/f ::LysM-Cre mice. Therefore, MAFB induces Ccl12, Arg1, c-Maf, and Il- 10. In Mafb f/f ::LysM-Cre mice, Ccl12 expression was 2.5-and 100-fold lower in wound tissue and bone marrow-derived M2 macrophages, respectively ( Figure 4B,C). Hence, Mafb induces Ccl12 expression in both wound tissue and macrophages in vitro.
In a previous study, Arg1 f/f ::Tie2-Cre mice presented with reduction in collagen accumulation and delayed wound healing [17]. Here, Masson's trichrome staining showed that collagen accumulation decreased on day 5 in Mafb f/f ::LysM-Cre mice ( Figure 1E). We confirmed Arg1 expression in day 3 and 5 granulation tissue using immunostaining ( Figure 4D). The ratios of Arg1-positive area per granulation tissue area were significantly reduced in the Mafb f/f ::LysM-Cre mice on days 3 and 5. The expression levels of the macrophage marker galectin-3 (Mac2) did not differ between Mafb f/f and Mafb f/f ::LysM-Cre mice (Supplementary Figure S3). Therefore, Arg1 expression may have specifically decreased in the macrophages that infiltrated the granulation tissue through Mafb deletion. Thus, MAFB may induce Ccl12, Ccl2, and Arg1 during wound healing.

Mafb-Positive Macrophages Show Heterogeneous Gene Expression
Both tissue-resident and infiltrated macrophages are intricately involved in wound healing [19]. We aimed to identify the cell population expressing genes that MAFB could regulate. To this end, we analyzed the data for the day 5 whole wound tissue, as well as for the macrophages infiltrating the wounds [20,21]. In the latter case, we traced the RFP + monocytes injected before harvesting the wound tissue ( Figure 4E,F).
Data for the day 4 and 14 wound macrophages were then analyzed (GEO accession no. GSE183489) ( Figure 4F) [21]. Mafb was not expressed in cluster 9 and the Ccl2 population resembled the Mafb population. The latter had a slightly higher expression level than that of c-Maf. However, most c-Maf + cells expressed Mafb mainly in clusters 2, 4, and 6 ( Figure 4F). Ccl12, Arg1, Il-10, and Cxcl10 exhibited different expression patterns. Ccl12 was slightly expressed in all clusters, whereas Arg1 was expressed mainly in clusters 1, 3, and 4 and Il-10 was expressed mainly in clusters 0, 2, and 6. Therefore, the gene populations were not identical and most of them expressed Mafb. These findings were similar to those for the whole wound tissue analysis ( Figure 4E). Ccl2 expression differed between the whole tissue and infiltrated wound macrophages. Ccl2 expression was relatively low in the Cd68 + macrophages of whole wound tissue ( Figure 4E). In the infiltrated wound macrophages, however, Ccl2 was comparatively upregulated ( Figure 4F). Thus, Ccl2 was expressed mainly by infiltrated macrophages in wounds. Additionally, because TLR2 has been reported to be important for the wound healing process [18] and its gene is coexpressed with MAFB in humans [5], we confirmed the expression of Tlr2 in mice wound scRNA-seq data. The expression pattern of Tlr2 was more similar to that of Mafb than that of Tlr4. This may suggest the possibility of a relationship between the TLR2 pathway and MAFB (Supplementary Figure S4C). The foregoing findings indicate that not all genes regulated by MAFB were expressed within the same population and that MAFB-expressing cells constituted a heterogeneous population. Thus, MAFB may affect wound healing via several genes.

Discussion
To the best of our knowledge, the present study is the first to investigate the role of MAFB in skin wound healing in vivo. At 3-5 days after the lesions were induced, the wounds were relatively larger and the collagen areas had decreased in the Mafb f/f ::LysM-Cre mice compared with that in the Mafb f/f mice. At this time point, the Ccl12, Ccl2, and Arg1 expression levels were lower in the wound tissue-and bone marrow-derived M2type macrophages (Figure 4B-D). Therefore, MAFB induces CCL12, CCL2, and Arg1 in the macrophages mediating tissue repair during wound healing. It was reported that CCL12 is a ligand of CCR2. While CCL12 occurs only in mice, it has high homology to human CCL2 [15,22,23]. Moore et al. showed that CCR2-mediated signaling recruited fibroblasts to the pulmonary alveoli in a lung fibrosis model. Moreover, pulmonary fibrosis was suppressed by neutralizing CCL12 but not CCL2 [15,24]. Here, we observed that collagen decreased in the Mafb f/f ::LysM-Cre mice on day 5 after wounding ( Figure 1E). Hence, CCL12 and CCL2 may promote fibrosis during wound healing. Mafb deletion in macrophages decreased the numbers of Gr-1 − , CD11b + , Ly6C + cells on day 5 ( Figure 3). Thus, MAFB may participate in monocyte/macrophage recruitment. CCR2 signaling induces Ly6C hi monocyte/macrophage recruitment to wound sites and promotes wound repair [25]. In a pulmonary fibrosis model, CCL12 was important for the recruitment of exudate macrophages and Ly6C hi monocytes to the lungs [23]. It is, therefore, possible that CCL12 and CCL2 were induced by MAFB on day 5 and promoted Ly6C hi monocyte/macrophage recruitment to the wound sites. Nevertheless, the initial influx of Ly6C hi monocytes/macrophages into the wound occurred on day 3 and there were no significant differences between the Mafb f/f and Mafb f/f ::LysM-Cre mice at that time. Thus, it is improbable that the observed decreases in the number of cells on day 5 had any significant effect on delayed wound healing.
Arg1 also contributed to the observed decrease in collagen in the Mafb f/f ::LysM-Cre mice. This hydrolytic enzyme degrades arginine into urea and ornithine. The latter may be converted to proline required for collagen synthesis and polyamines, which are essential for cell growth and differentiation. All three processes are vital to wound healing [17,26]. There were no significant differences between Mafb f/f and Mafb f/f ::LysM-Cre mice in terms of Arg1 mRNA expression in their wound tissues ( Figure 4B). However, an in vitro analysis of BM-derived M2 macrophages showed that Arg1 mRNA expression was significantly reduced in Mafb f/f ::LysM-Cre mice ( Figure 4C). Further, histological analysis showed that Arg1 expression was significantly decreased in the granulation tissues of Mafb f/f ::LysM-Cre mice on days 3 and 5 ( Figure 4D). Arg1 is also expressed at high levels in keratinocytes and especially in those at wound edges [27]. The edges of the wound tissues used in RT-qPCR ( Figure 4B) probably contained keratinocytes. For this reason, Arg1 mRNA expression might not have been rigorously evaluated in the granulation tissue. The foregoing results suggest that MAFB may induce Arg1 and, by extension, collagen production in granulation tissue.
The mRNA expression levels of c-Maf and Il-10 did not differ in the wound tissue. Nevertheless, Mafb deletion downregulated c-Maf and Il-10 in in vitro BM-derived M2 macrophages. Keratinocytes, T cells, and monocytes/macrophages express c-Maf [28,29]. T cells and monocytes/macrophages produce IL-10 [30]. Keratinocytes produce IL-10 after wounding [30]. RT-qPCR analysis of wound tissue ( Figure 4B) may therefore detect c-Maf and Il-10 induced by cells other than macrophages in the granulation tissue. In the present study, we found that MAFB induces c-Maf whereas M-CSF, IL-4, and IL-13 induce Il-10 in M2 macrophages. As c-Maf might promote IL-10 expression in macrophages [31], MAFB could regulate Il-10 expression via c-Maf regulation.
An evaluation of previously reported single-cell RNA sequencing (scRNA-Seq) data on whole wound tissue and infiltrated wound macrophages revealed the relationship between Mafb and its downstream genes. The scRNA-Seq data disclosed that Mafb is expressed in most clusters whereas the other cytokines are only expressed in a few of them. The preceding data indicate that Mafb + macrophages are highly heterogeneous. The roles of Mafb + macrophages may vary depending on their relative gene expression levels. We found that macrophage MAFB induces C1q [7]. C1q is a recognition signal that triggers the classical complement pathway, promotes the clearance of apoptotic cells, and contributes to cytokine/chemokine induction, depending on the receptors involved [32]. C1qa −/− mice presented with defective vascularization. However, local C1q application rescued this process [33]. These discoveries indicate that MAFB may contribute to angiogenesis and cytokine/chemokine induction and that it has functional diversity. The findings of the present work suggest that MAFB participates in wound closure, collagen production, and monocyte/macrophage recruitment by regulating the expression of Ccl12, Ccl2, Arg1, and other genes in various macrophage populations. In this manner, MAFB promotes integrated wound healing. Future studies on MAFB functions in each cluster will help elucidate the roles of macrophages in wound healing through scRNA-seq using Mafbdeficient wound tissue.
Previously, we showed that MAFB is regulated by nuclear receptor transcription factors. In vitro results show that agonists of PPARs, LXRs, RARs, and glucocorticoid receptors (GRs) increase MAFB expression in macrophages [7,9]. These agonists have been shown to induce immunosuppression and fibrosis during focal segmental glomerulosclerosis (FSGS), and MAFB may be an important hub in these signals [34]. However, we could not confirm whether these agonists are also functional during wound healing in this study. Further, Mafb −/− mice embryos showed impaired epidermal differentiation, indicating the importance of MAFB in keratinocytes for differentiation of keratinocytes [35]. In this study, we found that macrophage-specific, Mafb-deficient mice also showed impaired epithelialization, but the mechanism remains unclear. Furthermore, the roles of MAFB in other kinds of injuries, such as burns or wounds in other organs, are unknown. This study is the first step toward showing the potential of MAFB as a new therapeutic target. Further analysis of the transcriptional regulation of the MAFB gene may provide support for the applicability of MAFB for tissue repair under several conditions and in the field of plastic surgery.

Mice
Mafb GFP knock-in mice (Mafb heterozygous (Mafb gfp/+ )) with a C57BL/6J background were used in the present study. Mouse genomic DNA was extracted from the tail tips and the genotypes were characterized using PCR. The sequences of the genotyping primers were previously reported [9]. For the Mafb conditional knockout mice, Mafb was flanked by a loxP element with a neomycin-resistant gene via homologous recombination in C57BL/6 background ES cells [36]. Mafb f/f mice were mated with LysM-Cre mice (Jackson Laboratory, Hancock, ME, USA) to delete Mafb in the macrophage lineage. Mafb expression was under the control of the endogenous Lyz2 promoter [37]. The mice were maintained under specific pathogen-free conditions in a laboratory animal resource center at the University of Tsukuba, Japan.

Wounding and Wound Collection
The back hair of male mice aged 8-10 weeks was shaved and the mice were anesthetized with isoflurane. Two full-thickness wounds were made ≥5 mm apart on the denuded back skin with a 5 mm diameter biopsy punch (Kai Industries Co., Ltd., Gifu, Japan, Cat. No. BP-50F). Analyses were performed on 3, 5, 7, and 10 d after wounding. The mice were sacrificed by CO 2 inhalation and the wounds were collected with an 8 mm diameter biopsy punch (Kai Industries Co., Ltd., Gifu, Japan, Cat. No. BP-80F).

Wound Area Quantification
The wound areas were evaluated on days 0, 3, 5, 7, and 10 after the injuries were inflicted to determine the wound closure rates. Rulers were positioned adjacent to the wounds and photographs were taken and analyzed using Photoshop software (https://www.adobe.com/products/photoshop.html, accessed on 17 August 2022). Data were expressed as percentages of the initial wound area.

Histological Analysis
Wound tissues were fixed in Mildform 10N

qRT-PCR
Total RNA was collected with an Isogen Kit (Nippon Gene, Tokyo, Japan, Cat. No. 311-02501). The cDNA was synthesized with a QuantiTect Reverse Transcription Kit (Qiagen, Hilden, Germany, Cat. No. 205313). Gene expression levels were determined using RT-PCR performed on a Thermal Cycler Dice Real Time System Single TP850 (Takara Bio Inc., Shiga, Japan) and a THUNDERBIRD ® SYBR ® qPCR Mix (TOYOBO Co., Ltd., Osaka, Japan, Cat. No. QPS-201). The mRNA levels were normalized to that of Hprt. The primer nucleotide sequences are shown in Supplementary Table S1.

Analysis of scRNA-Seq Data
Single-nuclei RNA-Seq data for wound and non-wound macrophages were analyzed. CD45 − and CD31 − cells were isolated using flow cytometry from dorsal skin wound beds on day 5. The characteristics of cells with CD68 expression > 0 were then determined. Data processing and analysis were performed with Scanpy (https://github.com/scverse/scanpy, accessed on 17 August 2022). UMAP (https://github.com/lmcinnes/umap accessed on 17 August 2022) and gene expression analyses were subsequently performed. The UMI count was the expression matrix unit (Shook et al., GSE140512) [20]. The raw singlenuclei RNA-Seq data were downloaded from Gene Expression Omnibus (GEO) and then processed with Scanpy. Scanpy performs normalizations and analyzes and generates figures. Cells with <200 or >2500 unique feature counts and >5% mitochondrial counts were filtered out. After the quality control, the cells were subjected to downstream analysis. They were log-normalized and their highly variable genes were identified and subjected to a principal component analysis (PCA), which reduces dimensionality and reveals the main variations in data. A neighborhood graph was then computed, plotted with UMAP, and colored with Leiden. Genes associated with wound healing were selected and their expression levels and distribution were displayed in the macrophage clusters.
Single-nuclei RNA-Seq of mouse wound macrophages was performed. Wounds collected 4 or 14 d after the injury and monocyte transfusion on days 2 or 12 were pooled and analyzed using single-nuclei RNA-Seq analysis. The raw single-nuclei RNA-Seq data were obtained from GEO and processed with Scanpy. Scanpy performs log-normalization and highly variable genes were identified and subjected to PCA. A neighborhood graph was then computed, plotted with UMAP, and colored with Leiden. Genes associated with wound healing were selected and their expression levels and distribution were displayed in the macrophage clusters (Willenborg et al., GSE183489) [21].

Statistical Analyses
Data are means ± S.E.M. Significant differences between group pairs were analyzed by Student's t-test using Microsoft Excel. Differences were considered statistically significant at p < 0.05.

Institutional Review Board Statement:
All experiments were performed in compliance with the relevant Japanese and institutional laws and guidelines and were approved by the University of Tsukuba Animal Ethics Committee under Authorization No. 21-174.

Data Availability Statement:
The data presented in this article or in Supplementary Materials are available from the corresponding author upon reasonable request.

Conflicts of Interest:
The authors declare no conflict of interest.